{"generatedAt":"2026-09-09T01:21:20.762Z","snapshotCapturedAt":"2026-09-08T17:08:58.223Z","internalCapturedAt":"2026-09-08T00:00:00.000Z","methodology":"https://opencharts.com/benchmarks/methodology","attribution":[{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks","license":"CC BY 4.0"},{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset","license":"CC BY 4.0"},{"publisher":"OpenCharts internal evaluation","url":"https://opencharts.com/benchmarks/methodology#internal","license":"Published by OpenCharts"}],"benchmarks":[{"slug":"gpqa-diamond","name":"GPQA Diamond","category":"reasoning","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/benchmarks/gpqa-diamond","scale":"percent","provenance":"epoch-run","chanceLevel":0.25,"itemCount":198,"caveat":"Frontier models cluster in the eighties and nineties; Epoch's standard errors on this test are 2 to 3 points.","hardSet":false,"frontier":{"value":0.9577020202020202,"display":"95.8%"},"url":"https://opencharts.com/benchmarks/tests/gpqa-diamond"},{"slug":"hle","name":"Humanity's Last Exam","category":"reasoning","publisher":"Center for AI Safety & Scale AI","publisherUrl":"https://lastexam.ai","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":2500,"caveat":null,"hardSet":true,"frontier":{"value":0.465,"display":"46.5%"},"url":"https://opencharts.com/benchmarks/tests/hle"},{"slug":"arc-agi-2","name":"ARC-AGI-2","category":"reasoning","publisher":"ARC Prize Foundation","publisherUrl":"https://arcprize.org/leaderboard","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":120,"caveat":"Scores depend on the compute budget a lab chose; ARC Prize publishes cost per task beside every score and this ranking does not.","hardSet":true,"frontier":{"value":0.95,"display":"95.0%"},"url":"https://opencharts.com/benchmarks/tests/arc-agi-2"},{"slug":"arc-agi","name":"ARC-AGI-1","category":"reasoning","publisher":"ARC Prize Foundation","publisherUrl":"https://arcprize.org/leaderboard","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":100,"caveat":"Close to saturated at the frontier; scores depend on the compute budget a lab chose.","hardSet":false,"frontier":{"value":0.985,"display":"98.5%"},"url":"https://opencharts.com/benchmarks/tests/arc-agi"},{"slug":"livebench","name":"LiveBench","category":"reasoning","publisher":"LiveBench","publisherUrl":"https://livebench.ai","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":"Epoch's copy of this leaderboard lags the live site, so few current models have a score here.","hardSet":false,"frontier":{"value":82.35,"display":"82.3%"},"url":"https://opencharts.com/benchmarks/tests/livebench"},{"slug":"simplebench","name":"SimpleBench","category":"reasoning","publisher":"SimpleBench / LM Council","publisherUrl":"https://simple-bench.com","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":0.8190000000000001,"display":"81.9%"},"url":"https://opencharts.com/benchmarks/tests/simplebench"},{"slug":"critpt","name":"CritPt","category":"reasoning","publisher":"CritPt / Artificial Analysis","publisherUrl":"https://artificialanalysis.ai/evaluations/critpt","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":0.323,"display":"32.3%"},"url":"https://opencharts.com/benchmarks/tests/critpt"},{"slug":"simpleqa-verified","name":"SimpleQA Verified","category":"knowledge","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/benchmarks/simpleqa-verified","scale":"percent","provenance":"epoch-run","chanceLevel":null,"itemCount":1000,"caveat":null,"hardSet":false,"frontier":{"value":0.756,"display":"75.6%"},"url":"https://opencharts.com/benchmarks/tests/simpleqa-verified"},{"slug":"swe-bench-verified","name":"SWE-bench Verified","category":"coding","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/benchmarks/swe-bench-verified","scale":"percent","provenance":"epoch-run","chanceLevel":null,"itemCount":500,"caveat":"One shared scaffold for every model; labs' own numbers with bespoke scaffolds can run 10 points higher and are not used here.","hardSet":true,"frontier":{"value":0.8347107438016529,"display":"83.5%"},"url":"https://opencharts.com/benchmarks/tests/swe-bench-verified"},{"slug":"aider-polyglot","name":"Aider Polyglot","category":"coding","publisher":"Aider","publisherUrl":"https://aider.chat/docs/leaderboards/","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":225,"caveat":null,"hardSet":false,"frontier":{"value":88,"display":"88.0%"},"url":"https://opencharts.com/benchmarks/tests/aider-polyglot"},{"slug":"scicode","name":"SciCode","category":"coding","publisher":"SciCode / Artificial Analysis","publisherUrl":"https://artificialanalysis.ai/evaluations/scicode","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":338,"caveat":null,"hardSet":false,"frontier":{"value":0.62037037037037,"display":"62.0%"},"url":"https://opencharts.com/benchmarks/tests/scicode"},{"slug":"weirdml","name":"WeirdML","category":"coding","publisher":"Håvard Ihle","publisherUrl":"https://htihle.github.io/weirdml.html","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":0.929,"display":"92.9%"},"url":"https://opencharts.com/benchmarks/tests/weirdml"},{"slug":"frontiercode","name":"FrontierCode","category":"coding","publisher":"Cognition","publisherUrl":"https://epoch.ai/benchmarks/frontiercode","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":"Published by Cognition, a coding-agent vendor, from its own harness; not independently reproduced.","hardSet":false,"frontier":{"value":0.5348,"display":"53.5%"},"url":"https://opencharts.com/benchmarks/tests/frontiercode"},{"slug":"cursorbench","name":"CursorBench","category":"coding","publisher":"Cursor","publisherUrl":"https://epoch.ai/benchmarks/cursorbench","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":"Published by Cursor from its own product traffic and harness; not independently reproduced.","hardSet":false,"frontier":{"value":0.7340000000000001,"display":"73.4%"},"url":"https://opencharts.com/benchmarks/tests/cursorbench"},{"slug":"arena-webdev","name":"Code Arena (WebDev)","category":"coding","publisher":"Arena","publisherUrl":"https://arena.ai/leaderboard/webdev","scale":"elo","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":1797.0426317593628,"display":"1,797"},"url":"https://opencharts.com/benchmarks/tests/arena-webdev"},{"slug":"aime-2024-2025","name":"Mock AIME 2024–2025","category":"math","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","scale":"percent","provenance":"epoch-run","chanceLevel":null,"itemCount":45,"caveat":"45 problems: one problem is 2.2 points, so gaps of a few points are within noise.","hardSet":false,"frontier":{"value":1,"display":"100.0%"},"url":"https://opencharts.com/benchmarks/tests/aime-2024-2025"},{"slug":"math-level-5","name":"MATH Level 5","category":"math","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/benchmarks/math-level-5","scale":"percent","provenance":"epoch-run","chanceLevel":null,"itemCount":null,"caveat":"Epoch flags likely training contamination on this set; treat small gaps as meaningless.","hardSet":false,"frontier":{"value":0.9813066465256798,"display":"98.1%"},"url":"https://opencharts.com/benchmarks/tests/math-level-5"},{"slug":"frontiermath","name":"FrontierMath (Tiers 1–3)","category":"math","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/frontiermath","scale":"percent","provenance":"epoch-run","chanceLevel":null,"itemCount":290,"caveat":"Commissioned by OpenAI, which has access to much of the problem set; Epoch discloses this and keeps a holdout it does not share.","hardSet":true,"frontier":{"value":0.936842105263158,"display":"93.7%"},"url":"https://opencharts.com/benchmarks/tests/frontiermath"},{"slug":"frontiermath-tier-4","name":"FrontierMath Tier 4","category":"math","publisher":"Epoch AI (internal runs)","publisherUrl":"https://epoch.ai/frontiermath","scale":"percent","provenance":"epoch-run","chanceLevel":null,"itemCount":48,"caveat":"48 problems: one problem is about 2 points. Same OpenAI commissioning disclosure as tiers 1 to 3.","hardSet":true,"frontier":{"value":0.976,"display":"97.6%"},"url":"https://opencharts.com/benchmarks/tests/frontiermath-tier-4"},{"slug":"proofbench","name":"ProofBench","category":"math","publisher":"Vals AI","publisherUrl":"https://epoch.ai/benchmarks/proofbench","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":1,"display":"100.0%"},"url":"https://opencharts.com/benchmarks/tests/proofbench"},{"slug":"terminal-bench","name":"Terminal-Bench","category":"agentic","publisher":"Terminal-Bench","publisherUrl":"https://www.tbench.ai/leaderboard","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":"Harnesses differ by model, so this compares model-plus-harness systems, not models alone.","hardSet":true,"frontier":{"value":0.8470000000000001,"display":"84.7%"},"url":"https://opencharts.com/benchmarks/tests/terminal-bench"},{"slug":"gdpval","name":"GDPval","category":"agentic","publisher":"OpenAI (external evaluations)","publisherUrl":"https://epoch.ai/benchmarks/gdpval","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":220,"caveat":"Authored and graded by OpenAI, which also competes on it. Treat as a lab-run leaderboard.","hardSet":true,"frontier":{"value":0.49700000000000005,"display":"49.7%"},"url":"https://opencharts.com/benchmarks/tests/gdpval"},{"slug":"metr-time-horizons","name":"METR Time Horizon","category":"agentic","publisher":"METR","publisherUrl":"https://metr.org/research","scale":"score","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":"METR publishes wide confidence intervals around each horizon; the point estimate is used here.","hardSet":true,"frontier":{"value":1044.780145,"display":"17.4 h"},"url":"https://opencharts.com/benchmarks/tests/metr-time-horizons"},{"slug":"osworld-2","name":"OSWorld 2.0","category":"agentic","publisher":"XLANG Lab","publisherUrl":"https://osworld-v2.xlang.ai","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":108,"caveat":"A separate, far harder benchmark from OSWorld-Verified: binary completion is low for every model, and the leaderboard also reports a partial-credit score.","hardSet":false,"frontier":{"value":0.3143,"display":"31.4%"},"url":"https://opencharts.com/benchmarks/tests/osworld-2"},{"slug":"apex-agents","name":"APEX-Agents","category":"agentic","publisher":"Mercor","publisherUrl":"https://epoch.ai/benchmarks/apex-agents","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":0.474,"display":"47.4%"},"url":"https://opencharts.com/benchmarks/tests/apex-agents"},{"slug":"vending-bench-2","name":"Vending-Bench 2","category":"agentic","publisher":"Andon Labs","publisherUrl":"https://andonlabs.com/evals/vending-bench-2","scale":"score","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":"Andon Labs estimates a strong human operator at roughly $63,000, so every model is far from the ceiling.","hardSet":false,"frontier":{"value":11181.87333333333,"display":"$11,182"},"url":"https://opencharts.com/benchmarks/tests/vending-bench-2"},{"slug":"cybench","name":"Cybench","category":"agentic","publisher":"Stanford","publisherUrl":"https://cybench.github.io","scale":"percent","provenance":"developer-reported","chanceLevel":null,"itemCount":40,"caveat":"Developer-reported: the numbers come from the labs' own model cards, collected by Epoch AI, not from an independent run.","hardSet":false,"frontier":{"value":0.93,"display":"93.0%"},"url":"https://opencharts.com/benchmarks/tests/cybench"},{"slug":"deep-research-bench","name":"DeepResearch Bench","category":"agentic","publisher":"DeepResearch Bench","publisherUrl":"https://deepresearch-bench.github.io","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":100,"caveat":null,"hardSet":false,"frontier":{"value":0.5529999999999999,"display":"55.3%"},"url":"https://opencharts.com/benchmarks/tests/deep-research-bench"},{"slug":"arena-vision","name":"Vision Arena","category":"multimodal","publisher":"Arena","publisherUrl":"https://arena.ai/leaderboard/vision","scale":"elo","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":1313.059170076326,"display":"1,313"},"url":"https://opencharts.com/benchmarks/tests/arena-vision"},{"slug":"arena-text","name":"Text Arena","category":"human-preference","publisher":"Arena","publisherUrl":"https://arena.ai/leaderboard/text","scale":"elo","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":1507.164171675996,"display":"1,507"},"url":"https://opencharts.com/benchmarks/tests/arena-text"},{"slug":"fiction-livebench","name":"Fiction.LiveBench","category":"long-context","publisher":"Fiction.live","publisherUrl":"https://fiction.live/stories/Fiction-liveBench","scale":"percent","provenance":"leaderboard","chanceLevel":null,"itemCount":null,"caveat":null,"hardSet":false,"frontier":{"value":1,"display":"100.0%"},"url":"https://opencharts.com/benchmarks/tests/fiction-livebench"}],"models":[{"slug":"gpt-6-astra","name":"GPT-6 Astra","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":1,"rankRange":{"min":1,"max":1},"index":99,"indexRange":{"min":99,"max":99.7},"grade":"A+","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":100,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-09-03","url":"https://opencharts.com/benchmarks/gpt-6-astra","categories":[{"category":"reasoning","mean":99.6,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":100,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":97.7,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":99.8,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":98.5,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9577020202020202,"display":"95.8%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.95,"display":"95.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.985,"display":"98.5%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"critpt","value":0.317142857142857,"display":"31.7%","normalized":98.2,"normalizedMeaning":"98.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.756,"display":"75.6%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.564814814814815,"display":"56.5%","normalized":91,"normalizedMeaning":"91.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.9287,"display":"92.9%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.5326,"display":"53.3%","normalized":99.6,"normalizedMeaning":"99.6% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · harness: codex · reasoning effort: max"},{"benchmark":"arena-webdev","value":1797.0426317593628,"display":"1,797","normalized":100,"normalizedMeaning":"best published result","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":1,"display":"100.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.936842105263158,"display":"93.7%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.976,"display":"97.6%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.99,"display":"99.0%","normalized":99,"normalizedMeaning":"99.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.467,"display":"46.7%","normalized":98.5,"normalizedMeaning":"98.5% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"claude-fable-5-1","name":"Claude Fable 5.1","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":2,"rankRange":{"min":2,"max":2},"index":97,"indexRange":{"min":96.6,"max":97.8},"grade":"A+","eligible":true,"verification":"independent","hardSetScore":95.3,"hardSetMean":95.3,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2026-09-01","url":"https://opencharts.com/benchmarks/claude-fable-5-1","categories":[{"category":"reasoning","mean":97.5,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":93.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":97,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":96.6,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":88.4,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":99.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"hle","value":0.465,"display":"46.5%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"arc-agi-2","value":0.9,"display":"90.0%","normalized":94.7,"normalizedMeaning":"94.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.975,"display":"97.5%","normalized":99,"normalizedMeaning":"99.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"critpt","value":0.311428571428571,"display":"31.1%","normalized":96.4,"normalizedMeaning":"96.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simpleqa-verified","value":0.708,"display":"70.8%","normalized":93.7,"normalizedMeaning":"93.7% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.62037037037037,"display":"62.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.929,"display":"92.9%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.5091,"display":"50.9%","normalized":95.2,"normalizedMeaning":"95.2% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium · harness: claude-code · reasoning effort: medium"},{"benchmark":"cursorbench","value":0.7340000000000001,"display":"73.4%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1762.2464527581055,"display":"1,762","normalized":90,"normalizedMeaning":"45% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":1,"display":"100.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.9017543859649124,"display":"90.2%","normalized":96.3,"normalizedMeaning":"96.3% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.8780487804878049,"display":"87.8%","normalized":90,"normalizedMeaning":"90.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":1,"display":"100.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"apex-agents","value":0.474,"display":"47.4%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":5421.556666666666,"display":"$5,422","normalized":76.7,"normalizedMeaning":"76.7% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1504.213300810029,"display":"1,504","normalized":99.2,"normalizedMeaning":"50% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-02"}]},{"slug":"claude-fable-5","name":"Claude Fable 5","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":3,"rankRange":{"min":3,"max":4},"index":92.7,"indexRange":{"min":91.9,"max":95.6},"grade":"A","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":93.1,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-06-09","url":"https://opencharts.com/benchmarks/claude-fable-5","categories":[{"category":"reasoning","mean":93.7,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":93.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":89.4,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":95.1,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":86.6,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":100,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":100,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8585858585858586,"display":"85.9%","normalized":86,"normalizedMeaning":"86.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FS5pm7ctk9zgPfbyHrbD9AV.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.8916666666666667,"display":"89.2%","normalized":93.9,"normalizedMeaning":"93.9% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.985,"display":"98.5%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simplebench","value":0.8190000000000001,"display":"81.9%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"critpt","value":0.2857142857,"display":"28.6%","normalized":88.5,"normalizedMeaning":"88.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.707,"display":"70.7%","normalized":93.5,"normalizedMeaning":"93.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"scicode","value":0.601851851851852,"display":"60.2%","normalized":97,"normalizedMeaning":"97.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.9194,"display":"91.9%","normalized":99,"normalizedMeaning":"99.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.5348,"display":"53.5%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: claude-code · reasoning effort: xhigh"},{"benchmark":"cursorbench","value":0.705,"display":"70.5%","normalized":96,"normalizedMeaning":"96.0% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1629.0458934793564,"display":"1,629","normalized":55.1,"normalizedMeaning":"28% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":1,"display":"100.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FhwrnvtEbDAocFtjcsHqsQh.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.8701754385964913,"display":"87.0%","normalized":92.9,"normalizedMeaning":"92.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FE28uZiJvrPgKtawbJpoZQ8.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.902,"display":"90.2%","normalized":92.4,"normalizedMeaning":"92.4% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FReDAtJGGfRtrEGGWf4Zbsd.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.95,"display":"95.0%","normalized":95,"normalizedMeaning":"95.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"apex-agents","value":0.45,"display":"45.0%","normalized":94.9,"normalizedMeaning":"94.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":5680.258,"display":"$5,680","normalized":78.2,"normalizedMeaning":"78.2% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arena-vision","value":1313.059170076326,"display":"1,313","normalized":100,"normalizedMeaning":"best published result","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1507.164171675996,"display":"1,507","normalized":100,"normalizedMeaning":"best published result","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"claude-opus-5","name":"Claude Opus 5","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":4,"rankRange":{"min":3,"max":4},"index":92.6,"indexRange":{"min":91.7,"max":94.3},"grade":"A","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":87.2,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-24","url":"https://opencharts.com/benchmarks/claude-opus-5","categories":[{"category":"reasoning","mean":96,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":79.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":90.6,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":91.1,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":97.3,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":93.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":96,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9387626262626264,"display":"93.9%","normalized":97.3,"normalizedMeaning":"97.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.9042,"display":"90.4%","normalized":95.2,"normalizedMeaning":"95.2% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.975,"display":"97.5%","normalized":99,"normalizedMeaning":"99.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simplebench","value":0.8059999999999999,"display":"80.6%","normalized":98.4,"normalizedMeaning":"98.4% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.291428571,"display":"29.1%","normalized":90.2,"normalizedMeaning":"90.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.599,"display":"59.9%","normalized":79.2,"normalizedMeaning":"79.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.556712962962963,"display":"55.7%","normalized":89.7,"normalizedMeaning":"89.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.9178,"display":"91.8%","normalized":98.8,"normalizedMeaning":"98.8% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.5338,"display":"53.4%","normalized":99.8,"normalizedMeaning":"99.8% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · harness: claude-code · reasoning effort: medium"},{"benchmark":"cursorbench","value":0.7000000000000001,"display":"70.0%","normalized":95.4,"normalizedMeaning":"95.4% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1687.610196102199,"display":"1,688","normalized":69.5,"normalizedMeaning":"35% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9888888888888888,"display":"98.9%","normalized":98.9,"normalizedMeaning":"98.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.856140350877193,"display":"85.6%","normalized":91.4,"normalizedMeaning":"91.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fff7GeS7rdperjxSBZ7MHZY.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.7317073170731707,"display":"73.2%","normalized":75,"normalizedMeaning":"75.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fji3kMn49Jcr9t4whxcpjj4.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.99,"display":"99.0%","normalized":99,"normalizedMeaning":"99.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"osworld-2","value":0.3143,"display":"31.4%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"apex-agents","value":0.435,"display":"43.5%","normalized":91.8,"normalizedMeaning":"91.8% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"vending-bench-2","value":11181.87333333333,"display":"$11,182","normalized":100,"normalizedMeaning":"best published result","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1290.3531193024714,"display":"1,290","normalized":93.5,"normalizedMeaning":"47% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1493.3276603832862,"display":"1,493","normalized":96,"normalizedMeaning":"48% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gpt-5-6-sol","name":"GPT-5.6 Sol","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":5,"rankRange":{"min":5,"max":5},"index":89.8,"indexRange":{"min":88.7,"max":92.4},"grade":"A-","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":92.5,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-09","url":"https://opencharts.com/benchmarks/gpt-5-6-sol","categories":[{"category":"reasoning","mean":94.5,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":92.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":84,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":90.8,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":88.8,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":91.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":93.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9349747474747476,"display":"93.5%","normalized":96.8,"normalizedMeaning":"96.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.925,"display":"92.5%","normalized":97.4,"normalizedMeaning":"97.4% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.975,"display":"97.5%","normalized":99,"normalizedMeaning":"99.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simplebench","value":0.648,"display":"64.8%","normalized":79.1,"normalizedMeaning":"79.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.323,"display":"32.3%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.697,"display":"69.7%","normalized":92.2,"normalizedMeaning":"92.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.569444444444444,"display":"56.9%","normalized":91.8,"normalizedMeaning":"91.8% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.8876000000000001,"display":"88.8%","normalized":95.5,"normalizedMeaning":"95.5% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.4749,"display":"47.5%","normalized":88.8,"normalizedMeaning":"88.8% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: codex · reasoning effort: max"},{"benchmark":"cursorbench","value":0.672,"display":"67.2%","normalized":91.6,"normalizedMeaning":"91.6% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1617.4231420406804,"display":"1,617","normalized":52.5,"normalizedMeaning":"26% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · codex-harness · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":1,"display":"100.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.8912280701754386,"display":"89.1%","normalized":95.1,"normalizedMeaning":"95.1% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FVBpT2WWYKBZtUxG9Z3svGh.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.8292682926829268,"display":"82.9%","normalized":85,"normalizedMeaning":"85.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FQJ5rJMVypB4PMxPcBGftc8.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.8300000000000001,"display":"83.0%","normalized":83,"normalizedMeaning":"83.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"osworld-2","value":0.27340000000000003,"display":"27.3%","normalized":87,"normalizedMeaning":"87.0% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"apex-agents","value":0.399,"display":"39.9%","normalized":84.2,"normalizedMeaning":"84.2% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"vending-bench-2","value":9619.368,"display":"$9,619","normalized":95.2,"normalizedMeaning":"95.2% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1282.4198540102234,"display":"1,282","normalized":91.2,"normalizedMeaning":"46% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-08-27"},{"benchmark":"arena-text","value":1483.1510823374954,"display":"1,483","normalized":93.1,"normalizedMeaning":"47% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-09-02"}]},{"slug":"gpt-5-6-terra","name":"GPT-5.6 Terra","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":6,"rankRange":{"min":6,"max":6},"index":81.9,"indexRange":{"min":80.2,"max":85.3},"grade":"B+","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":84.2,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-09","url":"https://opencharts.com/benchmarks/gpt-5-6-terra","categories":[{"category":"reasoning","mean":87.1,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":57.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":74.1,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":84.5,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":86.5,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":86.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":88.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.933080808080808,"display":"93.3%","normalized":96.5,"normalizedMeaning":"96.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.839,"display":"83.9%","normalized":88.3,"normalizedMeaning":"88.3% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.965,"display":"96.5%","normalized":98,"normalizedMeaning":"98.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simplebench","value":0.489,"display":"48.9%","normalized":59.7,"normalizedMeaning":"59.7% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.3,"display":"30.0%","normalized":92.9,"normalizedMeaning":"92.9% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.432,"display":"43.2%","normalized":57.1,"normalizedMeaning":"57.1% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.539351851851852,"display":"53.9%","normalized":86.9,"normalizedMeaning":"86.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.7827,"display":"78.3%","normalized":84.3,"normalizedMeaning":"84.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.4131,"display":"41.3%","normalized":77.2,"normalizedMeaning":"77.2% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: codex · reasoning effort: max"},{"benchmark":"cursorbench","value":0.649,"display":"64.9%","normalized":88.4,"normalizedMeaning":"88.4% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1519.6694903416983,"display":"1,520","normalized":33.7,"normalizedMeaning":"17% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · codex-harness · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9972222222222222,"display":"99.7%","normalized":99.7,"normalizedMeaning":"99.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.8596491228070176,"display":"86.0%","normalized":91.8,"normalizedMeaning":"91.8% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJ6evjTMgHkrun8txnEvzdN.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.7073170731707317,"display":"70.7%","normalized":72.5,"normalizedMeaning":"72.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FZUygpkXqg3NjchXXfedAWV.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.74,"display":"74.0%","normalized":74,"normalizedMeaning":"74.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · reasoning effort: xhigh"},{"benchmark":"vending-bench-2","value":7343.206,"display":"$7,343","normalized":86.5,"normalizedMeaning":"86.5% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1265.728894053052,"display":"1,266","normalized":86.5,"normalizedMeaning":"43% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-08-27"},{"benchmark":"arena-text","value":1466.4202858826702,"display":"1,466","normalized":88.3,"normalizedMeaning":"44% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-09-02"}]},{"slug":"gpt-5-5","name":"GPT-5.5","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":7,"rankRange":{"min":6,"max":9},"index":80.2,"indexRange":{"min":78.8,"max":82.8},"grade":"B+","eligible":true,"verification":"independent","hardSetScore":88.7,"hardSetMean":88.7,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-23","url":"https://opencharts.com/benchmarks/gpt-5-5","categories":[{"category":"reasoning","mean":89.4,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":83.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":74.8,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":74.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":81.5,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":92.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":92.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9065656565656566,"display":"90.7%","normalized":92.8,"normalizedMeaning":"92.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2CdomqtBVz2it7a2Zx2G2t.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"low"},{"benchmark":"arc-agi-2","value":0.85,"display":"85.0%","normalized":89.5,"normalizedMeaning":"89.5% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"arc-agi","value":0.95,"display":"95.0%","normalized":96.4,"normalizedMeaning":"96.4% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simplebench","value":0.6900000000000001,"display":"69.0%","normalized":84.2,"normalizedMeaning":"84.2% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lmcouncil.ai/benchmarks","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.271428571428571,"display":"27.1%","normalized":84,"normalizedMeaning":"84.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simpleqa-verified","value":0.63,"display":"63.0%","normalized":83.3,"normalizedMeaning":"83.3% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTijCafTTJUFrJ6jPGvBbgq.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"scicode","value":0.561342592592593,"display":"56.1%","normalized":90.5,"normalizedMeaning":"90.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"weirdml","value":0.8491,"display":"84.9%","normalized":91.4,"normalizedMeaning":"91.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"frontiercode","value":0.4296,"display":"43.0%","normalized":80.3,"normalizedMeaning":"80.3% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: codex · reasoning effort: xhigh"},{"benchmark":"cursorbench","value":0.584,"display":"58.4%","normalized":79.6,"normalizedMeaning":"79.6% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · reasoning level: Extra High"},{"benchmark":"arena-webdev","value":1509.6175361396854,"display":"1,510","normalized":32.1,"normalizedMeaning":"16% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · codex-harness · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8444444444444444,"display":"84.4%","normalized":84.4,"normalizedMeaning":"84.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FCdmtwZ8NR8YeTXWiLVTRDJ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"low"},{"benchmark":"frontiermath","value":0.8526315789473684,"display":"85.3%","normalized":91,"normalizedMeaning":"91.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath-tier-4","value":0.725,"display":"72.5%","normalized":74.3,"normalizedMeaning":"74.3% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"proofbench","value":0.5,"display":"50.0%","normalized":50,"normalizedMeaning":"50.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · reasoning effort: xhigh"},{"benchmark":"terminal-bench","value":0.8470000000000001,"display":"84.7%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: NexAU-AHE"},{"benchmark":"osworld-2","value":0.13,"display":"13.0%","normalized":41.4,"normalizedMeaning":"41.4% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"apex-agents","value":0.385,"display":"38.5%","normalized":81.2,"normalizedMeaning":"81.2% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"vending-bench-2","value":7523.843999999999,"display":"$7,524","normalized":87.2,"normalizedMeaning":"87.2% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"deep-research-bench","value":0.54,"display":"54.0%","normalized":97.6,"normalizedMeaning":"97.6% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arena-vision","value":1286.0992704392563,"display":"1,286","normalized":92.3,"normalizedMeaning":"46% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1482.0850003935163,"display":"1,482","normalized":92.8,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"kimi-k3","name":"Kimi K3","lab":"Moonshot AI","country":{"code":"CN","name":"China"},"rank":8,"rankRange":{"min":6,"max":10},"index":79.6,"indexRange":{"min":77.2,"max":83.5},"grade":"B","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":60.2,"hardSetCount":3,"openWeights":true,"firstParty":false,"releaseDate":"2026-07-16","url":"https://opencharts.com/benchmarks/kimi-k3","categories":[{"category":"reasoning","mean":80.5,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":66.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":83,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":75.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":79,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":94.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9311868686868688,"display":"93.1%","normalized":96.3,"normalizedMeaning":"96.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.6041666666666666,"display":"60.4%","normalized":63.6,"normalizedMeaning":"63.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.945,"display":"94.5%","normalized":95.9,"normalizedMeaning":"95.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simplebench","value":0.6070000000000001,"display":"60.7%","normalized":74.1,"normalizedMeaning":"74.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"critpt","value":0.234,"display":"23.4%","normalized":72.4,"normalizedMeaning":"72.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.506,"display":"50.6%","normalized":66.9,"normalizedMeaning":"66.9% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmZLWt2JSYvJyGhMvji7CT8.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.586805555555556,"display":"58.7%","normalized":94.6,"normalizedMeaning":"94.6% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.8257,"display":"82.6%","normalized":88.9,"normalizedMeaning":"88.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.4417,"display":"44.2%","normalized":82.6,"normalizedMeaning":"82.6% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: mini-swe-agent · reasoning effort: none"},{"benchmark":"cursorbench","value":0.608,"display":"60.8%","normalized":82.8,"normalizedMeaning":"82.8% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1674.3489254278888,"display":"1,674","normalized":66.1,"normalizedMeaning":"33% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9722222222222222,"display":"97.2%","normalized":97.2,"normalizedMeaning":"97.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.721830985915493,"display":"72.2%","normalized":77,"normalizedMeaning":"77.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.3902439024390244,"display":"39.0%","normalized":40,"normalizedMeaning":"40.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FgZERSAQwxBAevQzBsWfUie.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.87,"display":"87.0%","normalized":87,"normalizedMeaning":"87.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.393,"display":"39.3%","normalized":82.9,"normalizedMeaning":"82.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":5165.035000000001,"display":"$5,165","normalized":75.1,"normalizedMeaning":"75.1% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1488.665886033902,"display":"1,489","normalized":94.7,"normalizedMeaning":"47% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-02"}]},{"slug":"claude-opus-4-8","name":"Claude Opus 4.8","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":9,"rankRange":{"min":7,"max":9},"index":79.5,"indexRange":{"min":77.7,"max":81.8},"grade":"B","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":72.9,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-05-28","url":"https://opencharts.com/benchmarks/claude-opus-4-8","categories":[{"category":"reasoning","mean":81.4,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":70.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":77.7,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":77.6,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":81.2,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":92,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":92.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9103535353535354,"display":"91.0%","normalized":93.3,"normalizedMeaning":"93.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.7208,"display":"72.1%","normalized":75.9,"normalizedMeaning":"75.9% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.925,"display":"92.5%","normalized":93.9,"normalizedMeaning":"93.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simplebench","value":0.648,"display":"64.8%","normalized":79.1,"normalizedMeaning":"79.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.20857142857142905,"display":"20.9%","normalized":64.6,"normalizedMeaning":"64.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.53,"display":"53.0%","normalized":70.1,"normalizedMeaning":"70.1% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FViC6pxQZUVS3E9M9iEubF7.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.534722222222222,"display":"53.5%","normalized":86.2,"normalizedMeaning":"86.2% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.8289,"display":"82.9%","normalized":89.2,"normalizedMeaning":"89.2% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"frontiercode","value":0.465,"display":"46.5%","normalized":86.9,"normalizedMeaning":"86.9% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: claude-code · reasoning effort: max"},{"benchmark":"cursorbench","value":0.623,"display":"62.3%","normalized":84.9,"normalizedMeaning":"84.9% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1562.7152909914344,"display":"1,563","normalized":41.2,"normalizedMeaning":"21% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9833333333333332,"display":"98.3%","normalized":98.3,"normalizedMeaning":"98.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.8,"display":"80.0%","normalized":85.4,"normalizedMeaning":"85.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FGVmgtgKMwEqc54f3jA7auP.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.5609756097560976,"display":"56.1%","normalized":57.5,"normalizedMeaning":"57.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJiqgqBTfkGx797uowZgJbk.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.6900000000000001,"display":"69.0%","normalized":69,"normalizedMeaning":"69.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"osworld-2","value":0.206,"display":"20.6%","normalized":65.5,"normalizedMeaning":"65.5% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"apex-agents","value":0.425,"display":"42.5%","normalized":89.7,"normalizedMeaning":"89.7% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"vending-bench-2","value":5787.428,"display":"$5,787","normalized":78.8,"normalizedMeaning":"78.8% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"deep-research-bench","value":0.502,"display":"50.2%","normalized":90.8,"normalizedMeaning":"90.8% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arena-vision","value":1285.3035434305757,"display":"1,285","normalized":92,"normalizedMeaning":"46% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1482.3133898091633,"display":"1,482","normalized":92.9,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gpt-5-4","name":"GPT-5.4","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":10,"rankRange":{"min":10,"max":11},"index":77.3,"indexRange":{"min":75.2,"max":79.1},"grade":"B","eligible":true,"verification":"independent","hardSetScore":80.3,"hardSetMean":80.3,"hardSetCount":7,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-05","url":"https://opencharts.com/benchmarks/gpt-5-4","categories":[{"category":"reasoning","mean":84,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":59.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":73.1,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":72,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":80.1,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":90.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":91.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.933,"display":"93.3%","normalized":96.5,"normalizedMeaning":"96.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FfFatyce8UvpN7ZivdmrhAy.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"hle","value":0.3624,"display":"36.2%","normalized":77.9,"normalizedMeaning":"77.9% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"arc-agi-2","value":0.7395,"display":"74.0%","normalized":77.8,"normalizedMeaning":"77.8% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"arc-agi","value":0.9367,"display":"93.7%","normalized":95.1,"normalizedMeaning":"95.1% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"critpt","value":0.23428571430000003,"display":"23.4%","normalized":72.5,"normalizedMeaning":"72.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"simpleqa-verified","value":0.451,"display":"45.1%","normalized":59.7,"normalizedMeaning":"59.7% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FAzCkm282PLhDdLJ8DnAQQc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"swe-bench-verified","value":0.768595041322314,"display":"76.9%","normalized":92.1,"normalizedMeaning":"92.1% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fi9HTMbAxcwsrTf2w2FuJ6N.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · high"},{"benchmark":"scicode","value":0.565972222222222,"display":"56.6%","normalized":91.2,"normalizedMeaning":"91.2% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"weirdml","value":0.777,"display":"77.7%","normalized":83.6,"normalizedMeaning":"83.6% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"arena-webdev","value":1462.7700439209696,"display":"1,463","normalized":25.5,"normalizedMeaning":"13% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · codex-harness · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9777777777777776,"display":"97.8%","normalized":97.8,"normalizedMeaning":"97.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FDpPahiEawokzgidE3cNZbS.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · high"},{"benchmark":"frontiermath","value":0.7859649122807018,"display":"78.6%","normalized":83.9,"normalizedMeaning":"83.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FPaYEQ37A2n6giYBwZGBnXY.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"frontiermath-tier-4","value":0.49,"display":"49.0%","normalized":50.2,"normalizedMeaning":"50.2% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FdoiQDghqnLzbE8CFijLNQJ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"proofbench","value":0.56,"display":"56.0%","normalized":56,"normalizedMeaning":"56.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh · reasoning effort: xhigh"},{"benchmark":"terminal-bench","value":0.818,"display":"81.8%","normalized":96.6,"normalizedMeaning":"96.6% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · agent: ForgeCode"},{"benchmark":"metr-time-horizons","value":341.735276,"display":"5.7 h","normalized":83.9,"normalizedMeaning":"83.9% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05"},{"benchmark":"apex-agents","value":0.36,"display":"36.0%","normalized":75.9,"normalizedMeaning":"75.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"vending-bench-2","value":6144.176,"display":"$6,144","normalized":80.7,"normalizedMeaning":"80.7% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05"},{"benchmark":"deep-research-bench","value":0.35100000000000003,"display":"35.1%","normalized":63.5,"normalizedMeaning":"63.5% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · low"},{"benchmark":"arena-vision","value":1280.7272861950855,"display":"1,281","normalized":90.7,"normalizedMeaning":"45% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1476.553434823906,"display":"1,477","normalized":91.2,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"claude-opus-4-6","name":"Claude Opus 4.6","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":11,"rankRange":{"min":10,"max":16},"index":75.4,"indexRange":{"min":72.6,"max":78.4},"grade":"B","eligible":true,"verification":"independent","hardSetScore":75.4,"hardSetMean":75.4,"hardSetCount":7,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-05","url":"https://opencharts.com/benchmarks/claude-opus-4-6","categories":[{"category":"reasoning","mean":83.5,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":62.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":66.6,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":60.6,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":91.1,"benchmarkCount":6,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":95.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":99.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9053030303030304,"display":"90.5%","normalized":92.6,"normalizedMeaning":"92.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FDiXCMhdfhcjaCUSzYWgACy.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"32k"},{"benchmark":"hle","value":0.3444,"display":"34.4%","normalized":74.1,"normalizedMeaning":"74.1% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.6917,"display":"69.2%","normalized":72.8,"normalizedMeaning":"72.8% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"120k"},{"benchmark":"arc-agi","value":0.94,"display":"94.0%","normalized":95.4,"normalizedMeaning":"95.4% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"120k"},{"benchmark":"simplebench","value":0.6759999999999999,"display":"67.6%","normalized":82.5,"normalizedMeaning":"82.5% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.47,"display":"47.0%","normalized":62.2,"normalizedMeaning":"62.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F85fhMxLaebWtLoi6oLpTbi.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"swe-bench-verified","value":0.7871900826446281,"display":"78.7%","normalized":94.3,"normalizedMeaning":"94.3% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2hH2GAPUFaJR7pfVKhhicg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"weirdml","value":0.7795000000000001,"display":"78.0%","normalized":83.9,"normalizedMeaning":"83.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.2664,"display":"26.6%","normalized":49.8,"normalizedMeaning":"49.8% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: claude-code · reasoning effort: high"},{"benchmark":"arena-webdev","value":1546.0107932226183,"display":"1,546","normalized":38.2,"normalizedMeaning":"19% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9444444444444444,"display":"94.4%","normalized":94.4,"normalizedMeaning":"94.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTUPC8VdX4jApFa4MGfNP4t.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"64k"},{"benchmark":"frontiermath","value":0.6596491228070176,"display":"66.0%","normalized":70.4,"normalizedMeaning":"70.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FkgJ2d6XbKj3dZutDcBYSKU.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.2682926829268293,"display":"26.8%","normalized":27.5,"normalizedMeaning":"27.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2hbbp8z6PhB7c2T33os9q5.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.5,"display":"50.0%","normalized":50,"normalizedMeaning":"50.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"terminal-bench","value":0.798,"display":"79.8%","normalized":94.2,"normalizedMeaning":"94.2% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: ForgeCode"},{"benchmark":"metr-time-horizons","value":718.80683,"display":"12.0 h","normalized":94.6,"normalizedMeaning":"94.6% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.324,"display":"32.4%","normalized":68.4,"normalizedMeaning":"68.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":8017.594,"display":"$8,018","normalized":89.3,"normalizedMeaning":"89.3% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"cybench","value":0.93,"display":"93.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www-cdn.anthropic.com/0dd865075ad3132672ee0ab40b05a53f14cf5288.pdf","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"}},{"benchmark":"deep-research-bench","value":0.5529999999999999,"display":"55.3%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arena-vision","value":1298.7626484538268,"display":"1,299","normalized":95.9,"normalizedMeaning":"48% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1504.8368533393275,"display":"1,505","normalized":99.3,"normalizedMeaning":"50% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gemini-3-7-flash","name":"Gemini 3.7 Flash","lab":"Google","country":{"code":"US","name":"United States"},"rank":12,"rankRange":{"min":9,"max":16},"index":75.2,"indexRange":{"min":71.9,"max":79.5},"grade":"B","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":67.6,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-08-13","url":"https://opencharts.com/benchmarks/gemini-3-7-flash","categories":[{"category":"reasoning","mean":82.2,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":91.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":76.2,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":67.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"human-preference","mean":95.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9482323232323232,"display":"94.8%","normalized":98.7,"normalizedMeaning":"98.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"arc-agi-2","value":0.8458333333333333,"display":"84.6%","normalized":89,"normalizedMeaning":"89.0% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.955,"display":"95.5%","normalized":97,"normalizedMeaning":"97.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"critpt","value":0.142857142857143,"display":"14.3%","normalized":44.2,"normalizedMeaning":"44.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simpleqa-verified","value":0.692,"display":"69.2%","normalized":91.5,"normalizedMeaning":"91.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F84ARuGyhQsjgh8umRmBvas.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"scicode","value":0.578703703703704,"display":"57.9%","normalized":93.3,"normalizedMeaning":"93.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"frontiercode","value":0.43590000000000007,"display":"43.6%","normalized":81.5,"normalizedMeaning":"81.5% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: chisel · reasoning effort: medium"},{"benchmark":"cursorbench","value":0.616,"display":"61.6%","normalized":83.9,"normalizedMeaning":"83.9% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · reasoning level: High"},{"benchmark":"arena-webdev","value":1586.8941808184122,"display":"1,587","normalized":46,"normalizedMeaning":"23% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9722222222222222,"display":"97.2%","normalized":97.2,"normalizedMeaning":"97.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.7157894736842105,"display":"71.6%","normalized":76.4,"normalizedMeaning":"76.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FaxMdgNDC8UYYLnZYLHZpF5.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath-tier-4","value":0.36585365853658536,"display":"36.6%","normalized":37.5,"normalizedMeaning":"37.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FKptKXK5nE2Ns2WbiaSSnbc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.58,"display":"58.0%","normalized":58,"normalizedMeaning":"58.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1490.926051738733,"display":"1,491","normalized":95.3,"normalizedMeaning":"48% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"grok-4-6","name":"Grok 4.6","lab":"xAI","country":{"code":"US","name":"United States"},"rank":13,"rankRange":{"min":10,"max":16},"index":75.2,"indexRange":{"min":71.2,"max":78.6},"grade":"B","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":57.8,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-08-12","url":"https://opencharts.com/benchmarks/grok-4-6","categories":[{"category":"reasoning","mean":82.1,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":65.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":80.2,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":63.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":90.1,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9400252525252524,"display":"94.0%","normalized":97.5,"normalizedMeaning":"97.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"arc-agi-2","value":0.6708333333333333,"display":"67.1%","normalized":70.6,"normalizedMeaning":"70.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"arc-agi","value":0.875,"display":"87.5%","normalized":88.8,"normalizedMeaning":"88.8% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"simplebench","value":0.7590000000000001,"display":"75.9%","normalized":92.7,"normalizedMeaning":"92.7% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.197142857142857,"display":"19.7%","normalized":61,"normalizedMeaning":"61.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simpleqa-verified","value":0.493,"display":"49.3%","normalized":65.2,"normalizedMeaning":"65.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTpHYtuPG6kYZFvMK7Z6Hqj.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"scicode","value":0.546296296296296,"display":"54.6%","normalized":88.1,"normalizedMeaning":"88.1% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"weirdml","value":0.6729,"display":"67.3%","normalized":72.4,"normalizedMeaning":"72.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.4801000000000001,"display":"48.0%","normalized":89.8,"normalizedMeaning":"89.8% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: grok-build · reasoning effort: high"},{"benchmark":"cursorbench","value":0.708,"display":"70.8%","normalized":96.5,"normalizedMeaning":"96.5% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · reasoning level: Extra High"},{"benchmark":"arena-webdev","value":1624.8562611967518,"display":"1,625","normalized":54.1,"normalizedMeaning":"27% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9916666666666668,"display":"99.2%","normalized":99.2,"normalizedMeaning":"99.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath","value":0.6596491228070176,"display":"66.0%","normalized":70.4,"normalizedMeaning":"70.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FZb2AKwa942D7dBTTcebAba.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath-tier-4","value":0.3170731707317073,"display":"31.7%","normalized":32.5,"normalizedMeaning":"32.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fbg3JRdd33iWfvRoGSvyXXE.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"proofbench","value":0.51,"display":"51.0%","normalized":51,"normalizedMeaning":"51.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.412,"display":"41.2%","normalized":86.9,"normalizedMeaning":"86.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":9047.0325,"display":"$9,047","normalized":93.2,"normalizedMeaning":"93.2% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"claude-opus-4-7","name":"Claude Opus 4.7","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":14,"rankRange":{"min":10,"max":16},"index":75.1,"indexRange":{"min":72.4,"max":77.8},"grade":"B","eligible":true,"verification":"independent","hardSetScore":76.6,"hardSetMean":76.6,"hardSetCount":6,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-16","url":"https://opencharts.com/benchmarks/claude-opus-4-7","categories":[{"category":"reasoning","mean":76.2,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":68.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":78.5,"benchmarkCount":6,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":64.8,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":80.9,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":96.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":98.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9015151515151516,"display":"90.2%","normalized":92.1,"normalizedMeaning":"92.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FVJezHrJ5YPCbpui8pZgcJg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"hle","value":0.362,"display":"36.2%","normalized":77.8,"normalizedMeaning":"77.8% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi-2","value":0.7583,"display":"75.8%","normalized":79.8,"normalizedMeaning":"79.8% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.935,"display":"93.5%","normalized":94.9,"normalizedMeaning":"94.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simplebench","value":0.617,"display":"61.7%","normalized":75.3,"normalizedMeaning":"75.3% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lmcouncil.ai/benchmarks","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.12,"display":"12.0%","normalized":37.2,"normalizedMeaning":"37.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.517,"display":"51.7%","normalized":68.4,"normalizedMeaning":"68.4% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2c4rvmf5N7QXTugrFLMZ83.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"swe-bench-verified","value":0.8347107438016529,"display":"83.5%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FnCmGWWMBip2s9AVEZNYXDX.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.545138888888889,"display":"54.5%","normalized":87.9,"normalizedMeaning":"87.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.7644,"display":"76.4%","normalized":82.3,"normalizedMeaning":"82.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.3854,"display":"38.5%","normalized":72.1,"normalizedMeaning":"72.1% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: claude-code · reasoning effort: max"},{"benchmark":"cursorbench","value":0.648,"display":"64.8%","normalized":88.3,"normalizedMeaning":"88.3% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1557.285079186198,"display":"1,557","normalized":40.2,"normalizedMeaning":"20% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.978,"display":"97.8%","normalized":97.8,"normalizedMeaning":"97.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjyCZnpXG2bh9G43X8pT75m.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath","value":0.7017543859649122,"display":"70.2%","normalized":74.9,"normalizedMeaning":"74.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FZkNQ7GER7Mc9FSPmpUzgKH.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.3170731707317073,"display":"31.7%","normalized":32.5,"normalizedMeaning":"32.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FP4LXQ9PyDDziEANMB5jVRd.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.54,"display":"54.0%","normalized":54,"normalizedMeaning":"54.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"terminal-bench","value":0.802,"display":"80.2%","normalized":94.7,"normalizedMeaning":"94.7% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: WOZCODE"},{"benchmark":"osworld-2","value":0.182,"display":"18.2%","normalized":57.9,"normalizedMeaning":"57.9% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"apex-agents","value":0.339,"display":"33.9%","normalized":71.5,"normalizedMeaning":"71.5% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"vending-bench-2","value":10936.763333333334,"display":"$10,937","normalized":99.3,"normalizedMeaning":"99.3% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1301.4995382886264,"display":"1,301","normalized":96.7,"normalizedMeaning":"48% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1502.1088204665782,"display":"1,502","normalized":98.5,"normalizedMeaning":"49% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gemini-3-1-pro","name":"Gemini 3.1 Pro","lab":"Google","country":{"code":"US","name":"United States"},"rank":15,"rankRange":{"min":11,"max":17},"index":73.4,"indexRange":{"min":69.9,"max":75.7},"grade":"B-","eligible":true,"verification":"independent","hardSetScore":77.6,"hardSetMean":77.6,"hardSetCount":7,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-19","url":"https://opencharts.com/benchmarks/gemini-3-1-pro","categories":[{"category":"reasoning","mean":88.4,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":97.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":71.6,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":53.2,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":80.5,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":90.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":94.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9444444444444444,"display":"94.4%","normalized":98.1,"normalizedMeaning":"98.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FBgmrBFsbsaD8rNpebv49Da.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview · high"},{"benchmark":"hle","value":0.4644,"display":"46.4%","normalized":99.9,"normalizedMeaning":"99.9% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arc-agi-2","value":0.771,"display":"77.1%","normalized":81.2,"normalizedMeaning":"81.2% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arc-agi","value":0.98,"display":"98.0%","normalized":99.5,"normalizedMeaning":"99.5% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"simplebench","value":0.7959999999999999,"display":"79.6%","normalized":97.2,"normalizedMeaning":"97.2% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"critpt","value":0.177142857142857,"display":"17.7%","normalized":54.8,"normalizedMeaning":"54.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"simpleqa-verified","value":0.735,"display":"73.5%","normalized":97.2,"normalizedMeaning":"97.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview · high"},{"benchmark":"swe-bench-verified","value":0.756198347107438,"display":"75.6%","normalized":90.6,"normalizedMeaning":"90.6% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F8QQQWDgmmEsmQVUJWcxx4P.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.58912037037037,"display":"58.9%","normalized":95,"normalizedMeaning":"95.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"weirdml","value":0.7206999999999999,"display":"72.1%","normalized":77.6,"normalizedMeaning":"77.6% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arena-webdev","value":1445.521203913088,"display":"1,446","normalized":23.4,"normalizedMeaning":"12% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.956,"display":"95.6%","normalized":95.6,"normalizedMeaning":"95.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbXgZZgzRE3wXoSRUgrbsVf.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"frontiermath","value":0.5964912280701754,"display":"59.6%","normalized":63.7,"normalizedMeaning":"63.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FhVifTUdQZrFEYabgCRu8Qq.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"frontiermath-tier-4","value":0.2682926829268293,"display":"26.8%","normalized":27.5,"normalizedMeaning":"27.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FeXVWr3RBeSsWy5DAU8ed46.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"proofbench","value":0.26,"display":"26.0%","normalized":26,"normalizedMeaning":"26.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.802,"display":"80.2%","normalized":94.7,"normalizedMeaning":"94.7% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · agent: TongAgents"},{"benchmark":"metr-time-horizons","value":384.147435,"display":"6.4 h","normalized":85.6,"normalizedMeaning":"85.6% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"apex-agents","value":0.335,"display":"33.5%","normalized":70.7,"normalizedMeaning":"70.7% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"vending-bench-2","value":3774.254,"display":"$3,774","normalized":65,"normalizedMeaning":"65.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"deep-research-bench","value":0.478,"display":"47.8%","normalized":86.4,"normalizedMeaning":"86.4% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · high"},{"benchmark":"arena-vision","value":1278.396272834183,"display":"1,278","normalized":90.1,"normalizedMeaning":"45% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-08-27"},{"benchmark":"arena-text","value":1486.7298381462967,"display":"1,487","normalized":94.1,"normalizedMeaning":"47% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-09-02"}]},{"slug":"gpt-5-6-luna","name":"GPT-5.6 Luna","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":16,"rankRange":{"min":11,"max":17},"index":72.9,"indexRange":{"min":70.6,"max":75.8},"grade":"B-","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":70.9,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-09","url":"https://opencharts.com/benchmarks/gpt-5-6-luna","categories":[{"category":"reasoning","mean":73.4,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":54.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":68.3,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":77.1,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":67.7,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":83,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":84.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9160353535353536,"display":"91.6%","normalized":94.1,"normalizedMeaning":"94.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.5954,"display":"59.5%","normalized":62.7,"normalizedMeaning":"62.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.88,"display":"88.0%","normalized":89.3,"normalizedMeaning":"89.3% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simplebench","value":0.468,"display":"46.8%","normalized":57.1,"normalizedMeaning":"57.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.206,"display":"20.6%","normalized":63.8,"normalizedMeaning":"63.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.41,"display":"41.0%","normalized":54.2,"normalizedMeaning":"54.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.525462962962963,"display":"52.5%","normalized":84.7,"normalizedMeaning":"84.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.6086,"display":"60.9%","normalized":65.5,"normalizedMeaning":"65.5% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.3981,"display":"39.8%","normalized":74.4,"normalizedMeaning":"74.4% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: codex · reasoning effort: max"},{"benchmark":"cursorbench","value":0.611,"display":"61.1%","normalized":83.2,"normalizedMeaning":"83.2% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1519.489089939797,"display":"1,519","normalized":33.7,"normalizedMeaning":"17% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · codex-harness · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9833333333333332,"display":"98.3%","normalized":98.3,"normalizedMeaning":"98.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.8210526315789474,"display":"82.1%","normalized":87.6,"normalizedMeaning":"87.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FYLAk5J3Wd8EruVXee8bAQb.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.6097560975609756,"display":"61.0%","normalized":62.5,"normalizedMeaning":"62.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FNLiiJJSyA9wmg2FHENFpAq.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.6,"display":"60.0%","normalized":60,"normalizedMeaning":"60.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"vending-bench-2","value":4094.712,"display":"$4,095","normalized":67.7,"normalizedMeaning":"67.7% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1253.5215025944576,"display":"1,254","normalized":83,"normalizedMeaning":"42% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-08-27"},{"benchmark":"arena-text","value":1452.5993829802887,"display":"1,453","normalized":84.4,"normalizedMeaning":"42% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-09-02"}]},{"slug":"claude-sonnet-5","name":"Claude Sonnet 5","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":17,"rankRange":{"min":12,"max":18},"index":71,"indexRange":{"min":68,"max":75.2},"grade":"B-","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":50,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2026-06-30","url":"https://opencharts.com/benchmarks/claude-sonnet-5","categories":[{"category":"reasoning","mean":72.9,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":44.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":72.1,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":67.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":75.3,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":87,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":87.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9053030303030304,"display":"90.5%","normalized":92.6,"normalizedMeaning":"92.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"simplebench","value":0.606,"display":"60.6%","normalized":74,"normalizedMeaning":"74.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.168571428571429,"display":"16.9%","normalized":52.2,"normalizedMeaning":"52.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.337,"display":"33.7%","normalized":44.6,"normalizedMeaning":"44.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FLPo74T9L3CawnKX5LLUrSZ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.53587962962963,"display":"53.6%","normalized":86.4,"normalizedMeaning":"86.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.6878000000000001,"display":"68.8%","normalized":74,"normalizedMeaning":"74.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.42730000000000007,"display":"42.7%","normalized":79.9,"normalizedMeaning":"79.9% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: claude-code · reasoning effort: xhigh"},{"benchmark":"cursorbench","value":0.615,"display":"61.5%","normalized":83.8,"normalizedMeaning":"83.8% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1536.9121369341317,"display":"1,537","normalized":36.6,"normalizedMeaning":"18% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9472222222222222,"display":"94.7%","normalized":94.7,"normalizedMeaning":"94.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath","value":0.656140350877193,"display":"65.6%","normalized":70,"normalizedMeaning":"70.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FV6YuGpcu8MCbJNXKoJMA6b.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.2926829268292683,"display":"29.3%","normalized":30,"normalizedMeaning":"30.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbqiVinHXFcgyFGPstJwXJN.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.77,"display":"77.0%","normalized":77,"normalizedMeaning":"77.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"apex-agents","value":0.325,"display":"32.5%","normalized":68.6,"normalizedMeaning":"68.6% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":6377.701666666665,"display":"$6,378","normalized":81.9,"normalizedMeaning":"81.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1267.4658585948616,"display":"1,267","normalized":87,"normalizedMeaning":"44% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1462.365843902521,"display":"1,462","normalized":87.2,"normalizedMeaning":"44% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gemini-3-5-flash","name":"Gemini 3.5 Flash","lab":"Google","country":{"code":"US","name":"United States"},"rank":18,"rankRange":{"min":17,"max":22},"index":68.2,"indexRange":{"min":63.7,"max":71.5},"grade":"C+","eligible":true,"verification":"independent","hardSetScore":66.4,"hardSetMean":66.4,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2026-05-19","url":"https://opencharts.com/benchmarks/gemini-3-5-flash","categories":[{"category":"reasoning","mean":80,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":87.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":69.3,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":55.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":76.6,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":91.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":92,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.928030303030303,"display":"92.8%","normalized":95.8,"normalizedMeaning":"95.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FdYYtcwENFRtm9jZTDA2p9C.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"arc-agi-2","value":0.7208,"display":"72.1%","normalized":75.9,"normalizedMeaning":"75.9% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.925,"display":"92.5%","normalized":93.9,"normalizedMeaning":"93.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simplebench","value":0.767,"display":"76.7%","normalized":93.7,"normalizedMeaning":"93.7% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.131428571428571,"display":"13.1%","normalized":40.7,"normalizedMeaning":"40.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simpleqa-verified","value":0.662,"display":"66.2%","normalized":87.6,"normalizedMeaning":"87.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FGJ6YsSuzyRD9YXe6SBwvsu.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"swe-bench-verified","value":0.7933884297520661,"display":"79.3%","normalized":95,"normalizedMeaning":"95.0% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FdwRzFGTghYEy4HNZajJXoc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"scicode","value":0.53125,"display":"53.1%","normalized":85.6,"normalizedMeaning":"85.6% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.6264000000000001,"display":"62.6%","normalized":67.4,"normalizedMeaning":"67.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"cursorbench","value":0.498,"display":"49.8%","normalized":67.8,"normalizedMeaning":"67.8% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1500.4139255270989,"display":"1,500","normalized":30.7,"normalizedMeaning":"15% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9555555555555556,"display":"95.6%","normalized":95.6,"normalizedMeaning":"95.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FeYmgjYJfM7Utm46nfoRfZ6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.6280701754385964,"display":"62.8%","normalized":67,"normalizedMeaning":"67.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F72DQVwbYiQW5pjhshjBaGU.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath-tier-4","value":0.2682926829268293,"display":"26.8%","normalized":27.5,"normalizedMeaning":"27.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FcXn5zzns6VFwbTgotDwZjV.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.31,"display":"31.0%","normalized":31,"normalizedMeaning":"31.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · reasoning effort: high"},{"benchmark":"vending-bench-2","value":5396.418333333334,"display":"$5,396","normalized":76.6,"normalizedMeaning":"76.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1284.2961073040717,"display":"1,284","normalized":91.7,"normalizedMeaning":"46% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1479.433788845838,"display":"1,479","normalized":92,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gpt-5-2","name":"GPT-5.2","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":19,"rankRange":{"min":17,"max":25},"index":67.7,"indexRange":{"min":63,"max":72},"grade":"C+","eligible":true,"verification":"independent","hardSetScore":71.2,"hardSetMean":71.2,"hardSetCount":8,"openWeights":false,"firstParty":false,"releaseDate":"2025-12-11","url":"https://opencharts.com/benchmarks/gpt-5-2","categories":[{"category":"reasoning","mean":70.5,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":49.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":62.1,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":53.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":78.6,"benchmarkCount":6,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":80.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":80.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.914,"display":"91.4%","normalized":93.8,"normalizedMeaning":"93.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FEjo6Fy7XYxUDcCwbejfcYX.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"hle","value":0.278,"display":"27.8%","normalized":59.8,"normalizedMeaning":"59.8% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11"},{"benchmark":"arc-agi-2","value":0.5291,"display":"52.9%","normalized":55.7,"normalizedMeaning":"55.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"arc-agi","value":0.862,"display":"86.2%","normalized":87.5,"normalizedMeaning":"87.5% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"simplebench","value":0.458,"display":"45.8%","normalized":55.9,"normalizedMeaning":"55.9% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · high"},{"benchmark":"simpleqa-verified","value":0.371,"display":"37.1%","normalized":49.1,"normalizedMeaning":"49.1% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FE3325ytkSNYP8ez53VA4fE.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"swe-bench-verified","value":0.737603305785124,"display":"73.8%","normalized":88.4,"normalizedMeaning":"88.4% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F9owdbpnWJ5pkQQC3GWKP5L.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · high"},{"benchmark":"weirdml","value":0.7219,"display":"72.2%","normalized":77.7,"normalizedMeaning":"77.7% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"arena-webdev","value":1416.3882847978043,"display":"1,416","normalized":20.1,"normalizedMeaning":"10% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9611111111111112,"display":"96.1%","normalized":96.1,"normalizedMeaning":"96.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FLiUMZyNSM8q93KDubWgYCh.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · high"},{"benchmark":"frontiermath","value":0.674,"display":"67.4%","normalized":71.9,"normalizedMeaning":"71.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"frontiermath-tier-4","value":0.317,"display":"31.7%","normalized":32.5,"normalizedMeaning":"32.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"proofbench","value":0.15,"display":"15.0%","normalized":15,"normalizedMeaning":"15.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · xhigh · reasoning effort: xhigh"},{"benchmark":"terminal-bench","value":0.649,"display":"64.9%","normalized":76.6,"normalizedMeaning":"76.6% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · medium · agent: Droid"},{"benchmark":"gdpval","value":0.49700000000000005,"display":"49.7%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · none"},{"benchmark":"metr-time-horizons","value":352.249302,"display":"5.9 h","normalized":84.4,"normalizedMeaning":"84.4% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · high"},{"benchmark":"apex-agents","value":0.344,"display":"34.4%","normalized":72.6,"normalizedMeaning":"72.6% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"vending-bench-2","value":3591.3259999999996,"display":"$3,591","normalized":63.4,"normalizedMeaning":"63.4% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11"},{"benchmark":"deep-research-bench","value":0.411,"display":"41.1%","normalized":74.3,"normalizedMeaning":"74.3% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · low"},{"benchmark":"arena-vision","value":1244.0574071496333,"display":"1,244","normalized":80.4,"normalizedMeaning":"40% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1437.8444873812937,"display":"1,438","normalized":80.3,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"deepseek-v4-pro","name":"DeepSeek V4 Pro","lab":"DeepSeek","country":{"code":"CN","name":"China"},"rank":20,"rankRange":{"min":17,"max":23},"index":67.5,"indexRange":{"min":63.3,"max":71.2},"grade":"C+","eligible":true,"verification":"independent","hardSetScore":63.5,"hardSetMean":63.5,"hardSetCount":4,"openWeights":true,"firstParty":false,"releaseDate":"2026-08-13","url":"https://opencharts.com/benchmarks/deepseek-v4-pro","categories":[{"category":"reasoning","mean":76.6,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":70,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":64.6,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":61.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":60.6,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":86.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9166666666666666,"display":"91.7%","normalized":94.2,"normalizedMeaning":"94.2% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.6125,"display":"61.3%","normalized":64.5,"normalizedMeaning":"64.5% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.905,"display":"90.5%","normalized":91.9,"normalizedMeaning":"91.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"low"},{"benchmark":"critpt","value":0.18,"display":"18.0%","normalized":55.7,"normalizedMeaning":"55.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.5290581162324649,"display":"52.9%","normalized":70,"normalizedMeaning":"70.0% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2xfFf6KZAp2gUWYhHGkFgs.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"swe-bench-verified","value":0.7763975155279503,"display":"77.6%","normalized":93,"normalizedMeaning":"93.0% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FWUk3f44JhxtAJaSxi6gFZR.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.5,"display":"50.0%","normalized":80.6,"normalizedMeaning":"80.6% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.6622,"display":"66.2%","normalized":71.3,"normalizedMeaning":"71.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.1764,"display":"17.6%","normalized":33,"normalizedMeaning":"33.0% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none · harness: mini-swe-agent · reasoning effort: none"},{"benchmark":"arena-webdev","value":1581.681919303248,"display":"1,582","normalized":44.9,"normalizedMeaning":"22% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high-20260813 · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9861111111111112,"display":"98.6%","normalized":98.6,"normalizedMeaning":"98.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.6456140350877193,"display":"64.6%","normalized":68.9,"normalizedMeaning":"68.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fb6yWmeFgsWeEiG9Qzs3NSf.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.2682926829268293,"display":"26.8%","normalized":27.5,"normalizedMeaning":"27.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fa4uV8HMMBAcHXLVSmZfBgX.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.5,"display":"50.0%","normalized":50,"normalizedMeaning":"50.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":3284.519999999999,"display":"$3,285","normalized":60.6,"normalizedMeaning":"60.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1459.6107477448018,"display":"1,460","normalized":86.4,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high-20260813 · board 2026-09-02"}]},{"slug":"grok-4-5","name":"Grok 4.5","lab":"xAI","country":{"code":"US","name":"United States"},"rank":21,"rankRange":{"min":21,"max":28},"index":64.2,"indexRange":{"min":59.3,"max":67.4},"grade":"C","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":47.1,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-08","url":"https://opencharts.com/benchmarks/grok-4-5","categories":[{"category":"reasoning","mean":74.8,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":63.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":64.1,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":53.7,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":69.1,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":91.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":89.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9343434343434344,"display":"93.4%","normalized":96.7,"normalizedMeaning":"96.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"arc-agi-2","value":0.5264,"display":"52.6%","normalized":55.4,"normalizedMeaning":"55.4% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.8717,"display":"87.2%","normalized":88.5,"normalizedMeaning":"88.5% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"simplebench","value":0.7000000000000001,"display":"70.0%","normalized":85.5,"normalizedMeaning":"85.5% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.154285714285714,"display":"15.4%","normalized":47.8,"normalizedMeaning":"47.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simpleqa-verified","value":0.483,"display":"48.3%","normalized":63.9,"normalizedMeaning":"63.9% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F7HWW3sjRstKzb9JpcnvLaV.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"scicode","value":0.540509259259259,"display":"54.1%","normalized":87.1,"normalizedMeaning":"87.1% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.4643,"display":"46.4%","normalized":50,"normalizedMeaning":"50.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"frontiercode","value":0.4244,"display":"42.4%","normalized":79.4,"normalizedMeaning":"79.4% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: grok-build · reasoning effort: high"},{"benchmark":"arena-webdev","value":1555.8721675772772,"display":"1,556","normalized":39.9,"normalizedMeaning":"20% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9777777777777776,"display":"97.8%","normalized":97.8,"normalizedMeaning":"97.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.5719298245614035,"display":"57.2%","normalized":61,"normalizedMeaning":"61.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath-tier-4","value":0.24390243902439024,"display":"24.4%","normalized":25,"normalizedMeaning":"25.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.31,"display":"31.0%","normalized":31,"normalizedMeaning":"31.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · reasoning effort: high"},{"benchmark":"apex-agents","value":0.342,"display":"34.2%","normalized":72.2,"normalizedMeaning":"72.2% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":3887.4259999999986,"display":"$3,887","normalized":66,"normalizedMeaning":"66.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1282.181707148235,"display":"1,282","normalized":91.1,"normalizedMeaning":"46% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1471.0801583876907,"display":"1,471","normalized":89.7,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"claude-sonnet-4-6","name":"Claude Sonnet 4.6","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":22,"rankRange":{"min":18,"max":28},"index":63.9,"indexRange":{"min":61,"max":69.7},"grade":"C","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":72.2,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-17","url":"https://opencharts.com/benchmarks/claude-sonnet-4-6","categories":[{"category":"reasoning","mean":62.3,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":47,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":63.8,"benchmarkCount":6,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":65.4,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":65.6,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":89.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":90,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8737373737373737,"display":"87.4%","normalized":88.1,"normalizedMeaning":"88.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F58eQmyCfPa3FufhXJFvAL2.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"32k"},{"benchmark":"arc-agi-2","value":0.6042,"display":"60.4%","normalized":63.6,"normalizedMeaning":"63.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.865,"display":"86.5%","normalized":87.8,"normalizedMeaning":"87.8% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"critpt","value":0.0314285714285714,"display":"3.1%","normalized":9.7,"normalizedMeaning":"9.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.355,"display":"35.5%","normalized":47,"normalizedMeaning":"47.0% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"swe-bench-verified","value":0.7520661157024794,"display":"75.2%","normalized":90.1,"normalizedMeaning":"90.1% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FKEdgiMmtJKHxgGiS82agYj.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.467592592592593,"display":"46.8%","normalized":75.4,"normalizedMeaning":"75.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.6607,"display":"66.1%","normalized":71.1,"normalizedMeaning":"71.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"frontiercode","value":0.24310000000000004,"display":"24.3%","normalized":45.5,"normalizedMeaning":"45.5% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: claude-code · reasoning effort: max"},{"benchmark":"cursorbench","value":0.49,"display":"49.0%","normalized":66.8,"normalizedMeaning":"66.8% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1521.4895156151356,"display":"1,521","normalized":34,"normalizedMeaning":"17% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.857936507936508,"display":"85.8%","normalized":85.8,"normalizedMeaning":"85.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F9LWdzVN5w4ihNvncsaCWMu.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"32k"},{"benchmark":"proofbench","value":0.45,"display":"45.0%","normalized":45,"normalizedMeaning":"45.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"terminal-bench","value":0.534,"display":"53.4%","normalized":63,"normalizedMeaning":"63.0% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Simplai Agent"},{"benchmark":"osworld-2","value":0.093,"display":"9.3%","normalized":29.6,"normalizedMeaning":"29.6% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"apex-agents","value":0.237,"display":"23.7%","normalized":50,"normalizedMeaning":"50.0% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"vending-bench-2","value":7204.139999999999,"display":"$7,204","normalized":85.9,"normalizedMeaning":"85.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"deep-research-bench","value":0.549,"display":"54.9%","normalized":99.3,"normalizedMeaning":"99.3% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arena-vision","value":1275.339422556073,"display":"1,275","normalized":89.2,"normalizedMeaning":"45% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1472.3891139817872,"display":"1,472","normalized":90,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"deepseek-v4-flash","name":"DeepSeek V4 Flash","lab":"DeepSeek","country":{"code":"CN","name":"China"},"rank":23,"rankRange":{"min":20,"max":28},"index":63.7,"indexRange":{"min":59.8,"max":67.5},"grade":"C","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":50.3,"hardSetCount":3,"openWeights":true,"firstParty":false,"releaseDate":"2026-07-31","url":"https://opencharts.com/benchmarks/deepseek-v4-flash","categories":[{"category":"reasoning","mean":74.8,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":44.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":57,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":59.2,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"human-preference","mean":80.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9103535353535354,"display":"91.0%","normalized":93.3,"normalizedMeaning":"93.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.6138888888888888,"display":"61.4%","normalized":64.6,"normalizedMeaning":"64.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arc-agi","value":0.89,"display":"89.0%","normalized":90.4,"normalizedMeaning":"90.4% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simplebench","value":0.611,"display":"61.1%","normalized":74.6,"normalizedMeaning":"74.6% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.165714285714286,"display":"16.6%","normalized":51.3,"normalizedMeaning":"51.3% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.33633633633633636,"display":"33.6%","normalized":44.5,"normalizedMeaning":"44.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTqFmiGvqozt7TSZYmhgF8m.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.49884259259259306,"display":"49.9%","normalized":80.4,"normalizedMeaning":"80.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.6296,"display":"63.0%","normalized":67.8,"normalizedMeaning":"67.8% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.188,"display":"18.8%","normalized":35.2,"normalizedMeaning":"35.2% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: chisel · reasoning effort: high"},{"benchmark":"arena-webdev","value":1579.6370978014602,"display":"1,580","normalized":44.5,"normalizedMeaning":"22% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9444444444444444,"display":"94.4%","normalized":94.4,"normalizedMeaning":"94.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.5754385964912281,"display":"57.5%","normalized":61.4,"normalizedMeaning":"61.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmKhVMEicZeGjdvJdTRd7mK.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.24390243902439024,"display":"24.4%","normalized":25,"normalizedMeaning":"25.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FWYs83xB6NwrkMNyQJtuMcQ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.56,"display":"56.0%","normalized":56,"normalizedMeaning":"56.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1438.4710745082564,"display":"1,438","normalized":80.5,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high-preview · board 2026-09-02"}]},{"slug":"glm-5-2","name":"GLM 5.2","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":24,"rankRange":{"min":21,"max":28},"index":63.3,"indexRange":{"min":59.6,"max":66.9},"grade":"C","eligible":true,"verification":"independent","hardSetScore":52.9,"hardSetMean":52.9,"hardSetCount":4,"openWeights":true,"firstParty":false,"releaseDate":"2026-06-16","url":"https://opencharts.com/benchmarks/glm-5-2","categories":[{"category":"reasoning","mean":66.6,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":45.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":69.6,"benchmarkCount":6,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":53.7,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":82.8,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":89.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.918560606060606,"display":"91.9%","normalized":94.5,"normalizedMeaning":"94.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"arc-agi-2","value":0.22775000000000004,"display":"22.8%","normalized":24,"normalizedMeaning":"24.0% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.77,"display":"77.0%","normalized":78.2,"normalizedMeaning":"78.2% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simplebench","value":0.588,"display":"58.8%","normalized":71.8,"normalizedMeaning":"71.8% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.20857142857142905,"display":"20.9%","normalized":64.6,"normalizedMeaning":"64.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.342,"display":"34.2%","normalized":45.2,"normalizedMeaning":"45.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FWgjFZzojpiv54XqUhvfuUG.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"swe-bench-verified","value":0.787,"display":"78.7%","normalized":94.3,"normalizedMeaning":"94.3% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F4AfhcmYVNrw6gM5u8CQsyy.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.50462962962963,"display":"50.5%","normalized":81.3,"normalizedMeaning":"81.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.7012,"display":"70.1%","normalized":75.5,"normalizedMeaning":"75.5% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"frontiercode","value":0.245,"display":"24.5%","normalized":45.8,"normalizedMeaning":"45.8% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none · harness: mini-swe-agent · reasoning effort: none"},{"benchmark":"cursorbench","value":0.55,"display":"55.0%","normalized":74.9,"normalizedMeaning":"74.9% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning level: Max"},{"benchmark":"arena-webdev","value":1586.5739929234114,"display":"1,587","normalized":45.9,"normalizedMeaning":"23% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8638888888888889,"display":"86.4%","normalized":86.4,"normalizedMeaning":"86.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.592057761732852,"display":"59.2%","normalized":63.2,"normalizedMeaning":"63.2% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.2926829268292683,"display":"29.3%","normalized":30,"normalizedMeaning":"30.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJebA8dYLVov6H49ZQSLXh5.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.35000000000000003,"display":"35.0%","normalized":35,"normalizedMeaning":"35.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: max"},{"benchmark":"apex-agents","value":0.35600000000000004,"display":"35.6%","normalized":75.1,"normalizedMeaning":"75.1% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":8313.778333333335,"display":"$8,314","normalized":90.5,"normalizedMeaning":"90.5% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1471.71832610685,"display":"1,472","normalized":89.8,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-02"}]},{"slug":"gemini-3-6-flash","name":"Gemini 3.6 Flash","lab":"Google","country":{"code":"US","name":"United States"},"rank":25,"rankRange":{"min":20,"max":28},"index":63.2,"indexRange":{"min":58.7,"max":67.5},"grade":"C","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":49.7,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-21","url":"https://opencharts.com/benchmarks/gemini-3-6-flash","categories":[{"category":"reasoning","mean":71.7,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":87.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":63.9,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":53.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"multimodal","mean":91.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":92.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9412878787878788,"display":"94.1%","normalized":97.7,"normalizedMeaning":"97.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"arc-agi-2","value":0.6041666666666666,"display":"60.4%","normalized":63.6,"normalizedMeaning":"63.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.9116666666666666,"display":"91.2%","normalized":92.6,"normalizedMeaning":"92.6% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"critpt","value":0.105714285714286,"display":"10.6%","normalized":32.7,"normalizedMeaning":"32.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simpleqa-verified","value":0.662,"display":"66.2%","normalized":87.6,"normalizedMeaning":"87.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FF83pa7Gfsb4UNVttTnS5Hk.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"scicode","value":0.52662037037037,"display":"52.7%","normalized":84.9,"normalizedMeaning":"84.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.561,"display":"56.1%","normalized":60.4,"normalizedMeaning":"60.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.3437,"display":"34.4%","normalized":64.3,"normalizedMeaning":"64.3% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: chisel · reasoning effort: medium"},{"benchmark":"cursorbench","value":0.535,"display":"53.5%","normalized":72.9,"normalizedMeaning":"72.9% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · reasoning level: High"},{"benchmark":"arena-webdev","value":1538.3722550096727,"display":"1,538","normalized":36.8,"normalizedMeaning":"18% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9416666666666668,"display":"94.2%","normalized":94.2,"normalizedMeaning":"94.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.5894736842105263,"display":"58.9%","normalized":62.9,"normalizedMeaning":"62.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fnb9LnDeeQyvRCqsbsTANGD.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath-tier-4","value":0.2195121951219512,"display":"22.0%","normalized":22.5,"normalizedMeaning":"22.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJCMjXU6FRiiLV3xvs9R2Xv.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.36,"display":"36.0%","normalized":36,"normalizedMeaning":"36.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning effort: high"},{"benchmark":"arena-vision","value":1284.8895872070825,"display":"1,285","normalized":91.9,"normalizedMeaning":"46% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1480.1982543453105,"display":"1,480","normalized":92.3,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"gemini-3-flash","name":"Gemini 3 Flash","lab":"Google","country":{"code":"US","name":"United States"},"rank":26,"rankRange":{"min":21,"max":30},"index":62.5,"indexRange":{"min":57,"max":66.6},"grade":"C","eligible":true,"verification":"independent","hardSetScore":54.8,"hardSetMean":54.8,"hardSetCount":5,"openWeights":false,"firstParty":false,"releaseDate":"2025-12-17","url":"https://opencharts.com/benchmarks/gemini-3-flash","categories":[{"category":"reasoning","mean":71.8,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":88.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":59.7,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":45.7,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":70.1,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":88.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":90.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8939393939393939,"display":"89.4%","normalized":91,"normalizedMeaning":"91.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FgVdraDT3pXmaExP9VbqME4.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview · high"},{"benchmark":"arc-agi-2","value":0.3361,"display":"33.6%","normalized":35.4,"normalizedMeaning":"35.4% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arc-agi","value":0.8467,"display":"84.7%","normalized":86,"normalizedMeaning":"86.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"simplebench","value":0.611,"display":"61.1%","normalized":74.6,"normalizedMeaning":"74.6% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"simpleqa-verified","value":0.668,"display":"66.8%","normalized":88.4,"normalizedMeaning":"88.4% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F7cc6cDCXjsPjtKt36bbeMR.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview · high"},{"benchmark":"swe-bench-verified","value":0.7541322314049587,"display":"75.4%","normalized":90.3,"normalizedMeaning":"90.3% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjxoyMNmTnYirjLJN6wRNmG.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"weirdml","value":0.616,"display":"61.6%","normalized":66.3,"normalizedMeaning":"66.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arena-webdev","value":1437.7238671086834,"display":"1,438","normalized":22.4,"normalizedMeaning":"11% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9555555555555556,"display":"95.6%","normalized":95.6,"normalizedMeaning":"95.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fa4uE6KgMhGSUjcnemRnbkA.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview · high"},{"benchmark":"frontiermath","value":0.512280701754386,"display":"51.2%","normalized":54.7,"normalizedMeaning":"54.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FEtgrDq6ei6kZe8K7QbXUe3.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"frontiermath-tier-4","value":0.17073170731707318,"display":"17.1%","normalized":17.5,"normalizedMeaning":"17.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FZczEhu69FQ9Cnns8NT6ew4.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"proofbench","value":0.15,"display":"15.0%","normalized":15,"normalizedMeaning":"15.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.643,"display":"64.3%","normalized":75.9,"normalizedMeaning":"75.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · agent: Junie CLI"},{"benchmark":"apex-agents","value":0.24,"display":"24.0%","normalized":50.6,"normalizedMeaning":"50.6% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"vending-bench-2","value":3634.715999999999,"display":"$3,635","normalized":63.8,"normalizedMeaning":"63.8% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"deep-research-bench","value":0.498,"display":"49.8%","normalized":90.1,"normalizedMeaning":"90.1% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · low"},{"benchmark":"arena-vision","value":1272.476964333666,"display":"1,272","normalized":88.4,"normalizedMeaning":"44% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1473.8268235299597,"display":"1,474","normalized":90.4,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"claude-opus-4-5","name":"Claude Opus 4.5","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":27,"rankRange":{"min":21,"max":30},"index":62.4,"indexRange":{"min":57.4,"max":66.4},"grade":"C","eligible":true,"verification":"independent","hardSetScore":59.4,"hardSetMean":59.4,"hardSetCount":8,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-24","url":"https://opencharts.com/benchmarks/claude-opus-4-5","categories":[{"category":"reasoning","mean":67.4,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":60.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":63.4,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":41,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":78.9,"benchmarkCount":7,"coveredCount":8,"indexed":true,"counted":true},{"category":"human-preference","mean":90.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.860479797979798,"display":"86.0%","normalized":86.3,"normalizedMeaning":"86.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FKwwo3UFcPDk6iLAeSeeRaa.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251101 · 32k"},{"benchmark":"hle","value":0.252,"display":"25.2%","normalized":54.2,"normalizedMeaning":"54.2% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101"},{"benchmark":"arc-agi-2","value":0.3764,"display":"37.6%","normalized":39.6,"normalizedMeaning":"39.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101 · 64k"},{"benchmark":"arc-agi","value":0.8,"display":"80.0%","normalized":81.2,"normalizedMeaning":"81.2% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101 · 64k"},{"benchmark":"simplebench","value":0.62,"display":"62.0%","normalized":75.7,"normalizedMeaning":"75.7% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101"},{"benchmark":"simpleqa-verified","value":0.457,"display":"45.7%","normalized":60.4,"normalizedMeaning":"60.4% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FCA5LW4phRnHSsZUGXEm8H6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251101 · 32k"},{"benchmark":"swe-bench-verified","value":0.7665289256198347,"display":"76.7%","normalized":91.8,"normalizedMeaning":"91.8% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmKQKxkpGDrFHBPBYR3mBrB.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251101"},{"benchmark":"weirdml","value":0.6374000000000001,"display":"63.7%","normalized":68.6,"normalizedMeaning":"68.6% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101 · 16k"},{"benchmark":"arena-webdev","value":1494.648600294498,"display":"1,495","normalized":29.8,"normalizedMeaning":"15% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101-high-32k · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8611111111111112,"display":"86.1%","normalized":86.1,"normalizedMeaning":"86.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FikUWdu6HJ2qRgZynfzDCFe.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251101 · 32k"},{"benchmark":"frontiermath","value":0.34385964912280703,"display":"34.4%","normalized":36.7,"normalizedMeaning":"36.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmS5EwCRZVpvUvJEehPUDZu.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251101 · 32k"},{"benchmark":"frontiermath-tier-4","value":0.04878048780487805,"display":"4.9%","normalized":5,"normalizedMeaning":"5.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fazk4theXxAsizmoi2HDD49.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251101 · 32k"},{"benchmark":"proofbench","value":0.36,"display":"36.0%","normalized":36,"normalizedMeaning":"36.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101 · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.631,"display":"63.1%","normalized":74.5,"normalizedMeaning":"74.5% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101 · agent: Droid"},{"benchmark":"gdpval","value":0.455,"display":"45.5%","normalized":91.5,"normalizedMeaning":"91.5% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101"},{"benchmark":"metr-time-horizons","value":292.994594,"display":"4.9 h","normalized":81.7,"normalizedMeaning":"81.7% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101"},{"benchmark":"apex-agents","value":0.207,"display":"20.7%","normalized":43.7,"normalizedMeaning":"43.7% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101"},{"benchmark":"vending-bench-2","value":4967.056666666668,"display":"$4,967","normalized":73.9,"normalizedMeaning":"73.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101"},{"benchmark":"cybench","value":0.82,"display":"82.0%","normalized":88.2,"normalizedMeaning":"88.2% of the best (93.0%)","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://assets.anthropic.com/m/64823ba7485345a7/Claude-Opus-4-5-System-Card.pdf","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"},"variantNote":"20251101"},{"benchmark":"deep-research-bench","value":0.5479999999999999,"display":"54.8%","normalized":99.1,"normalizedMeaning":"99.1% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101 · high"},{"benchmark":"arena-text","value":1472.8365910409484,"display":"1,473","normalized":90.2,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251101-high-32k · board 2026-09-02"}]},{"slug":"gpt-5","name":"GPT-5","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":28,"rankRange":{"min":21,"max":28},"index":62,"indexRange":{"min":59.4,"max":65},"grade":"C","eligible":true,"verification":"independent","hardSetScore":55,"hardSetMean":55,"hardSetCount":8,"openWeights":false,"firstParty":false,"releaseDate":"2025-08-07","url":"https://opencharts.com/benchmarks/gpt-5","categories":[{"category":"reasoning","mean":54.4,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":66.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":68.6,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":58.2,"benchmarkCount":5,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":66.7,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":71.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":79.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":96.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8617424242424242,"display":"86.2%","normalized":86.4,"normalizedMeaning":"86.4% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F9nPyNrZxwtoT7eS4DZbQF6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"hle","value":0.2532,"display":"25.3%","normalized":54.5,"normalizedMeaning":"54.5% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"arc-agi-2","value":0.0986,"display":"9.9%","normalized":10.4,"normalizedMeaning":"10.4% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"arc-agi","value":0.6567000000000001,"display":"65.7%","normalized":66.7,"normalizedMeaning":"66.7% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"simplebench","value":0.5670000000000001,"display":"56.7%","normalized":69.2,"normalizedMeaning":"69.2% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"critpt","value":0.126,"display":"12.6%","normalized":39,"normalizedMeaning":"39.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"simpleqa-verified","value":0.501,"display":"50.1%","normalized":66.3,"normalizedMeaning":"66.3% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FgTLurbyKU63i2opsGkJGwh.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"swe-bench-verified","value":0.7355371900826446,"display":"73.6%","normalized":88.1,"normalizedMeaning":"88.1% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FC85Bnci9xVzq5jmt8zNnbw.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"aider-polyglot","value":88,"display":"88.0%","normalized":100,"normalizedMeaning":"best published result","bestValue":88,"fieldSize":5,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://aider.chat/docs/leaderboards/#polyglot-leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"scicode","value":0.429398148148148,"display":"42.9%","normalized":69.2,"normalizedMeaning":"69.2% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07"},{"benchmark":"weirdml","value":0.6070000000000001,"display":"60.7%","normalized":65.3,"normalizedMeaning":"65.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"arena-webdev","value":1418.9679952087963,"display":"1,419","normalized":20.4,"normalizedMeaning":"10% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9138888888888888,"display":"91.4%","normalized":91.4,"normalizedMeaning":"91.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F3D8aaFBLZwQcnWHYKwvnmq.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"math-level-5","value":0.9813066465256798,"display":"98.1%","normalized":100,"normalizedMeaning":"best published result","bestValue":0.9813066465256798,"fieldSize":6,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/math-level-5","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"frontiermath","value":0.5543859649122806,"display":"55.4%","normalized":59.2,"normalizedMeaning":"59.2% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"frontiermath-tier-4","value":0.2195121951219512,"display":"22.0%","normalized":22.5,"normalizedMeaning":"22.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"proofbench","value":0.18,"display":"18.0%","normalized":18,"normalizedMeaning":"18.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.4960000000000001,"display":"49.6%","normalized":58.6,"normalizedMeaning":"58.6% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · agent: Codex CLI"},{"benchmark":"gdpval","value":0.348,"display":"34.8%","normalized":70,"normalizedMeaning":"70.0% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · medium"},{"benchmark":"metr-time-horizons","value":203.012577,"display":"3.4 h","normalized":76.4,"normalizedMeaning":"76.4% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"apex-agents","value":0.183,"display":"18.3%","normalized":38.6,"normalizedMeaning":"38.6% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"deep-research-bench","value":0.4960000000000001,"display":"49.6%","normalized":89.7,"normalizedMeaning":"89.7% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://drb.futuresearch.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · low"},{"benchmark":"arena-vision","value":1210.4111692189008,"display":"1,210","normalized":71.3,"normalizedMeaning":"36% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1434.3880306748788,"display":"1,434","normalized":79.4,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.969,"display":"96.9%","normalized":96.9,"normalizedMeaning":"96.9% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · medium"}]},{"slug":"kimi-k2-7-code","name":"Kimi K2.7 Code","lab":"Moonshot AI","country":{"code":"CN","name":"China"},"rank":29,"rankRange":{"min":29,"max":30},"index":58.6,"indexRange":{"min":56.2,"max":61.2},"grade":"C-","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":35.1,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2026-06-12","url":"https://opencharts.com/benchmarks/kimi-k2-7-code","categories":[{"category":"reasoning","mean":63.5,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":48.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":57.1,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":55.3,"benchmarkCount":3,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":66.4,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8787878787878788,"display":"87.9%","normalized":88.8,"normalizedMeaning":"88.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FXneEig6pACkAWeekvgCwNk.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simplebench","value":0.579,"display":"57.9%","normalized":70.7,"normalizedMeaning":"70.7% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.1,"display":"10.0%","normalized":31,"normalizedMeaning":"31.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.365,"display":"36.5%","normalized":48.3,"normalizedMeaning":"48.3% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F5AijYw5bFJtSuahnbzNo6Z.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.474537037037037,"display":"47.5%","normalized":76.5,"normalizedMeaning":"76.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.5412,"display":"54.1%","normalized":58.3,"normalizedMeaning":"58.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"frontiercode","value":0.3006,"display":"30.1%","normalized":56.2,"normalizedMeaning":"56.2% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: mini-swe-agent · reasoning effort: none"},{"benchmark":"cursorbench","value":0.49700000000000005,"display":"49.7%","normalized":67.7,"normalizedMeaning":"67.7% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1472.4504013663477,"display":"1,472","normalized":26.7,"normalizedMeaning":"13% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9555555555555556,"display":"95.6%","normalized":95.6,"normalizedMeaning":"95.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FENc5ZM5TpBYNKkTWN4xyQM.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.5403508771929825,"display":"54.0%","normalized":57.7,"normalizedMeaning":"57.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FX3RxRkt2pBUXYJjAQHsUo6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath-tier-4","value":0.12195121951219512,"display":"12.2%","normalized":12.5,"normalizedMeaning":"12.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FUTVSgirxQpTgmTHbgE9EUk.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"apex-agents","value":0.276,"display":"27.6%","normalized":58.2,"normalizedMeaning":"58.2% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":5082.936,"display":"$5,083","normalized":74.6,"normalizedMeaning":"74.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"glm-5-1","name":"GLM 5.1","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":30,"rankRange":{"min":29,"max":30},"index":57.5,"indexRange":{"min":54.7,"max":60.5},"grade":"C-","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":64.1,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2026-04-07","url":"https://opencharts.com/benchmarks/glm-5-1","categories":[{"category":"reasoning","mean":57.7,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":45,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":63.2,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":51.6,"benchmarkCount":3,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":77.9,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":88.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.898989898989899,"display":"89.9%","normalized":91.7,"normalizedMeaning":"91.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FCBbeMKRRSeRsXnvSfn7Png.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simplebench","value":0.551,"display":"55.1%","normalized":67.3,"normalizedMeaning":"67.3% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0457142857142857,"display":"4.6%","normalized":14.2,"normalizedMeaning":"14.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.34,"display":"34.0%","normalized":45,"normalizedMeaning":"45.0% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FR2fMFAtRXah864hMifdQGV.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"swe-bench-verified","value":0.7417,"display":"74.2%","normalized":88.9,"normalizedMeaning":"88.9% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/swe-bench-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.4375,"display":"43.8%","normalized":70.5,"normalizedMeaning":"70.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.5710000000000001,"display":"57.1%","normalized":61.5,"normalizedMeaning":"61.5% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1508.111304359068,"display":"1,508","normalized":31.9,"normalizedMeaning":"16% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9333333333333332,"display":"93.3%","normalized":93.3,"normalizedMeaning":"93.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fb7Lw9ExxAhVY3hdExoh4MP.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.3684210526315789,"display":"36.8%","normalized":39.3,"normalizedMeaning":"39.3% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FMhwme9odWJMzMjoYkW8r6B.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.22222000000000003,"display":"22.2%","normalized":22.2,"normalizedMeaning":"22.2% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":5634.406,"display":"$5,634","normalized":77.9,"normalizedMeaning":"77.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1466.0224725182957,"display":"1,466","normalized":88.2,"normalizedMeaning":"44% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"claude-sonnet-4-5","name":"Claude Sonnet 4.5","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":31,"rankRange":{"min":31,"max":32},"index":54.3,"indexRange":{"min":50.9,"max":57},"grade":"C-","eligible":true,"verification":"independent","hardSetScore":45.8,"hardSetMean":45.8,"hardSetCount":8,"openWeights":false,"firstParty":false,"releaseDate":"2025-09-29","url":"https://opencharts.com/benchmarks/claude-sonnet-4-5","categories":[{"category":"reasoning","mean":43.2,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":40.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":56.6,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":44.9,"benchmarkCount":5,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":72.5,"benchmarkCount":6,"coveredCount":8,"indexed":true,"counted":true},{"category":"human-preference","mean":85.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8232323232323232,"display":"82.3%","normalized":81,"normalizedMeaning":"81.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FXaGuWDxPtqVvyxqdMncKFj.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929 · 59k"},{"benchmark":"hle","value":0.13720000000000002,"display":"13.7%","normalized":29.5,"normalizedMeaning":"29.5% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929"},{"benchmark":"arc-agi-2","value":0.1361,"display":"13.6%","normalized":14.3,"normalizedMeaning":"14.3% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · 32k"},{"benchmark":"arc-agi","value":0.6367,"display":"63.7%","normalized":64.6,"normalizedMeaning":"64.6% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · 32k"},{"benchmark":"simplebench","value":0.543,"display":"54.3%","normalized":66.3,"normalizedMeaning":"66.3% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · 12k"},{"benchmark":"critpt","value":0.0114285714285714,"display":"1.1%","normalized":3.5,"normalizedMeaning":"3.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929"},{"benchmark":"simpleqa-verified","value":0.307,"display":"30.7%","normalized":40.6,"normalizedMeaning":"40.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FXQEGUQjpXupjdyKoGuSmum.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929 · 59k"},{"benchmark":"swe-bench-verified","value":0.7128099173553719,"display":"71.3%","normalized":85.4,"normalizedMeaning":"85.4% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FLti64yMEsBWSzgEMFCSVhp.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929"},{"benchmark":"scicode","value":0.446759259259259,"display":"44.7%","normalized":72,"normalizedMeaning":"72.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929"},{"benchmark":"weirdml","value":0.4771,"display":"47.7%","normalized":51.4,"normalizedMeaning":"51.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · 16k"},{"benchmark":"arena-webdev","value":1392.3567358310934,"display":"1,392","normalized":17.7,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929-high-32k · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.7777777777777778,"display":"77.8%","normalized":77.8,"normalizedMeaning":"77.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fk5PDL6MpCC6P5eLAhodJiS.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929 · 59k"},{"benchmark":"math-level-5","value":0.9773413897280968,"display":"97.7%","normalized":99.6,"normalizedMeaning":"99.6% of the best (98.1%)","bestValue":0.9813066465256798,"fieldSize":6,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/math-level-5","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929 · 32k"},{"benchmark":"frontiermath","value":0.23859649122807017,"display":"23.9%","normalized":25.5,"normalizedMeaning":"25.5% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F7oUkeLfwMK3MtkSNHX6QPu.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929 · 32k"},{"benchmark":"frontiermath-tier-4","value":0.024390243902439025,"display":"2.4%","normalized":2.5,"normalizedMeaning":"2.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fd3m9Xip8cof4s77z6LKkEh.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250929 · 32k"},{"benchmark":"proofbench","value":0.19,"display":"19.0%","normalized":19,"normalizedMeaning":"19.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929"},{"benchmark":"terminal-bench","value":0.465,"display":"46.5%","normalized":54.9,"normalizedMeaning":"54.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · agent: CAMEL-AI"},{"benchmark":"gdpval","value":0.425,"display":"42.5%","normalized":85.5,"normalizedMeaning":"85.5% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929"},{"benchmark":"metr-time-horizons","value":121.946588,"display":"2.0 h","normalized":69.1,"normalizedMeaning":"69.1% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · 16k"},{"benchmark":"vending-bench-2","value":3838.74,"display":"$3,839","normalized":65.6,"normalizedMeaning":"65.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929"},{"benchmark":"cybench","value":0.6,"display":"60.0%","normalized":64.5,"normalizedMeaning":"64.5% of the best (93.0%)","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://assets.anthropic.com/m/64823ba7485345a7/Claude-Opus-4-5-System-Card.pdf","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"},"variantNote":"20250929"},{"benchmark":"deep-research-bench","value":0.526,"display":"52.6%","normalized":95.1,"normalizedMeaning":"95.1% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://drb.futuresearch.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929 · 2k"},{"benchmark":"arena-text","value":1456.2681799330542,"display":"1,456","normalized":85.5,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250929-high-32k · board 2026-09-02"}]},{"slug":"gpt-5-1","name":"GPT-5.1","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":32,"rankRange":{"min":31,"max":32},"index":54,"indexRange":{"min":51.5,"max":56.4},"grade":"C-","eligible":true,"verification":"independent","hardSetScore":51.8,"hardSetMean":51.8,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-13","url":"https://opencharts.com/benchmarks/gpt-5-1","categories":[{"category":"reasoning","mean":52,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":63.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":58.6,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":88.6,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":51.3,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":82,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":85.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8762626262626263,"display":"87.6%","normalized":88.5,"normalizedMeaning":"88.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FCSVtMksEVjBNsGJGjW4GcQ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-11-13 · high"},{"benchmark":"hle","value":0.2368,"display":"23.7%","normalized":50.9,"normalizedMeaning":"50.9% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13"},{"benchmark":"arc-agi-2","value":0.1764,"display":"17.6%","normalized":18.6,"normalizedMeaning":"18.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · high"},{"benchmark":"arc-agi","value":0.7283,"display":"72.8%","normalized":73.9,"normalizedMeaning":"73.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · high"},{"benchmark":"livebench","value":78.79,"display":"78.8%","normalized":95.7,"normalizedMeaning":"95.7% of the best (82.3%)","bestValue":82.35,"fieldSize":1,"counted":false,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://livebench.ai/#/","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · high"},{"benchmark":"simplebench","value":0.532,"display":"53.2%","normalized":65,"normalizedMeaning":"65.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · high"},{"benchmark":"critpt","value":0.0485714285714286,"display":"4.9%","normalized":15,"normalizedMeaning":"15.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13"},{"benchmark":"simpleqa-verified","value":0.48,"display":"48.0%","normalized":63.5,"normalizedMeaning":"63.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FQCLnRy69NBH9ip3VYtQneh.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-11-13 · high"},{"benchmark":"swe-bench-verified","value":0.6797520661157025,"display":"68.0%","normalized":81.4,"normalizedMeaning":"81.4% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FYoHtZpMNdsRE6qdj2K4Ucj.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-11-13 · high"},{"benchmark":"scicode","value":0.43287037037037,"display":"43.3%","normalized":69.8,"normalizedMeaning":"69.8% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13"},{"benchmark":"weirdml","value":0.6077,"display":"60.8%","normalized":65.4,"normalizedMeaning":"65.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · high"},{"benchmark":"arena-webdev","value":1391.6887105657534,"display":"1,392","normalized":17.7,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8861111111111111,"display":"88.6%","normalized":88.6,"normalizedMeaning":"88.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmYKKTdV27TRJv9FpUg7rkn.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-11-13 · high"},{"benchmark":"terminal-bench","value":0.476,"display":"47.6%","normalized":56.2,"normalizedMeaning":"56.2% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · agent: Terminus 2"},{"benchmark":"apex-agents","value":0.17500000000000002,"display":"17.5%","normalized":36.9,"normalizedMeaning":"36.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · high"},{"benchmark":"vending-bench-2","value":1473.432,"display":"$1,473","normalized":34.8,"normalizedMeaning":"34.8% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13"},{"benchmark":"deep-research-bench","value":0.428,"display":"42.8%","normalized":77.4,"normalizedMeaning":"77.4% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-11-13 · low"},{"benchmark":"arena-vision","value":1249.9947390175118,"display":"1,250","normalized":82,"normalizedMeaning":"41% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1455.0471648717626,"display":"1,455","normalized":85.1,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"claude-opus-4-1","name":"Claude Opus 4.1","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":33,"rankRange":{"min":28,"max":34},"index":50.7,"indexRange":{"min":47.4,"max":62},"grade":"C-","eligible":true,"verification":"independent","hardSetScore":47.1,"hardSetMean":47.1,"hardSetCount":7,"openWeights":false,"firstParty":false,"releaseDate":"2025-08-05","url":"https://opencharts.com/benchmarks/claude-opus-4-1","categories":[{"category":"reasoning","mean":57.3,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":51.6,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":28.3,"benchmarkCount":3,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":66.6,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"human-preference","mean":83.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7727272727272727,"display":"77.3%","normalized":73.9,"normalizedMeaning":"73.9% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FerLgyKQs2mg2memE9SJvK5.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250805 · 16k"},{"benchmark":"hle","value":0.1152,"display":"11.5%","normalized":24.8,"normalizedMeaning":"24.8% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805"},{"benchmark":"simplebench","value":0.6,"display":"60.0%","normalized":73.3,"normalizedMeaning":"73.3% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805"},{"benchmark":"swe-bench-verified","value":0.7334710743801653,"display":"73.3%","normalized":87.9,"normalizedMeaning":"87.9% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FcA6Mq67yaynLhY8F5kes3c.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250805"},{"benchmark":"weirdml","value":0.4586,"display":"45.9%","normalized":49.4,"normalizedMeaning":"49.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805 · 16k"},{"benchmark":"arena-webdev","value":1388.9568714355896,"display":"1,389","normalized":17.4,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805 · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.6888888888888889,"display":"68.9%","normalized":68.9,"normalizedMeaning":"68.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FUH9RY8MCAotDgFocHKA4j6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250805 · 27k"},{"benchmark":"frontiermath","value":0.1263157894736842,"display":"12.6%","normalized":13.5,"normalizedMeaning":"13.5% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F5eRkmXCURH2pRbhJoukfnC.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250805 · 32k"},{"benchmark":"frontiermath-tier-4","value":0.024390243902439025,"display":"2.4%","normalized":2.5,"normalizedMeaning":"2.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F9zT72xRR6wZhiwraJTNPyz.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20250805 · 32k"},{"benchmark":"terminal-bench","value":0.38,"display":"38.0%","normalized":44.9,"normalizedMeaning":"44.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805 · agent: Terminus 2"},{"benchmark":"gdpval","value":0.436,"display":"43.6%","normalized":87.7,"normalizedMeaning":"87.7% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805"},{"benchmark":"metr-time-horizons","value":113.68507,"display":"114 min","normalized":68.1,"normalizedMeaning":"68.1% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805 · 16k"},{"benchmark":"cybench","value":0.42,"display":"42.0%","normalized":45.2,"normalizedMeaning":"45.2% of the best (93.0%)","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cybench.github.io/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"},"variantNote":"20250805"},{"benchmark":"deep-research-bench","value":0.483,"display":"48.3%","normalized":87.3,"normalizedMeaning":"87.3% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://drb.futuresearch.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805"},{"benchmark":"arena-text","value":1449.7073013438926,"display":"1,450","normalized":83.6,"normalizedMeaning":"42% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20250805-thinking-16k · board 2026-09-02"}]},{"slug":"gpt-5-4-mini","name":"GPT-5.4 mini","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":34,"rankRange":{"min":33,"max":34},"index":49.3,"indexRange":{"min":44.3,"max":53},"grade":"D","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":28.2,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-17","url":"https://opencharts.com/benchmarks/gpt-5-4-mini","categories":[{"category":"reasoning","mean":50.7,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":38.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":53.5,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":43.7,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":58.8,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":82.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":83.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8686868686868687,"display":"86.9%","normalized":87.4,"normalizedMeaning":"87.4% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FgKB2rCD7mHzFN2MCt2YCni.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"arc-agi-2","value":0.189,"display":"18.9%","normalized":19.9,"normalizedMeaning":"19.9% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"arc-agi","value":0.6367,"display":"63.7%","normalized":64.6,"normalizedMeaning":"64.6% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"critpt","value":0.1,"display":"10.0%","normalized":31,"normalizedMeaning":"31.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"simpleqa-verified","value":0.294,"display":"29.4%","normalized":38.9,"normalizedMeaning":"38.9% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FZ3pjKgAULY9j9wmzQpdqZG.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · high"},{"benchmark":"scicode","value":0.49884259259259306,"display":"49.9%","normalized":80.4,"normalizedMeaning":"80.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"weirdml","value":0.603,"display":"60.3%","normalized":64.9,"normalizedMeaning":"64.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · high"},{"benchmark":"frontiercode","value":0.2704,"display":"27.0%","normalized":50.6,"normalizedMeaning":"50.6% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · harness: codex · reasoning effort: xhigh"},{"benchmark":"arena-webdev","value":1397.1873915845283,"display":"1,397","normalized":18.2,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8888888888888888,"display":"88.9%","normalized":88.9,"normalizedMeaning":"88.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FQGMk6JNCYuUkFMF4zdVCsM.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"frontiermath","value":0.512280701754386,"display":"51.2%","normalized":54.7,"normalizedMeaning":"54.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"frontiermath-tier-4","value":0.0975609756097561,"display":"9.8%","normalized":10,"normalizedMeaning":"10.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"proofbench","value":0.21,"display":"21.0%","normalized":21,"normalizedMeaning":"21.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh · reasoning effort: xhigh"},{"benchmark":"apex-agents","value":0.246,"display":"24.6%","normalized":51.9,"normalizedMeaning":"51.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17"},{"benchmark":"deep-research-bench","value":0.363,"display":"36.3%","normalized":65.6,"normalizedMeaning":"65.6% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · low"},{"benchmark":"arena-vision","value":1252.068884827936,"display":"1,252","normalized":82.6,"normalizedMeaning":"41% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1448.201693060913,"display":"1,448","normalized":83.2,"normalizedMeaning":"42% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"inkling","name":"Inkling","lab":"Thinking Machines","country":{"code":"US","name":"United States"},"rank":35,"rankRange":{"min":35,"max":36},"index":42.8,"indexRange":{"min":36.5,"max":46.4},"grade":"D","eligible":true,"verification":"independent","hardSetScore":null,"hardSetMean":26.4,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-15","url":"https://opencharts.com/benchmarks/inkling","categories":[{"category":"reasoning","mean":57.3,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":53.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":38.7,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":32.4,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"human-preference","mean":80.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8825757575757576,"display":"88.3%","normalized":89.4,"normalizedMeaning":"89.4% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"arc-agi-2","value":0.3652777777777778,"display":"36.5%","normalized":38.5,"normalizedMeaning":"38.5% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.795,"display":"79.5%","normalized":80.7,"normalizedMeaning":"80.7% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simplebench","value":0.5,"display":"50.0%","normalized":61.1,"normalizedMeaning":"61.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0542857142857143,"display":"5.4%","normalized":16.8,"normalizedMeaning":"16.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simpleqa-verified","value":0.403,"display":"40.3%","normalized":53.3,"normalizedMeaning":"53.3% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbpnzC6CCi3hWuKCfqX4KEY.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"scicode","value":0.460648148148148,"display":"46.1%","normalized":74.3,"normalizedMeaning":"74.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"weirdml","value":0.3229,"display":"32.3%","normalized":34.8,"normalizedMeaning":"34.8% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"frontiercode","value":0.14,"display":"14.0%","normalized":26.2,"normalizedMeaning":"26.2% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: mini-swe-agent · reasoning effort: 0.99"},{"benchmark":"arena-webdev","value":1408.675879436946,"display":"1,409","normalized":19.3,"normalizedMeaning":"10% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8888888888888888,"display":"88.9%","normalized":88.9,"normalizedMeaning":"88.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath","value":0.3333333333333333,"display":"33.3%","normalized":35.6,"normalizedMeaning":"35.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F6KvQvuMMX5zAuteUc6pFcW.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath-tier-4","value":0.04878048780487805,"display":"4.9%","normalized":5,"normalizedMeaning":"5.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjvJPoRYkPH9Sb5MX252oHp.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"proofbench","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1439.4225598928058,"display":"1,439","normalized":80.7,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gemini-2-5-pro","name":"Gemini 2.5 Pro","lab":"Google","country":{"code":"US","name":"United States"},"rank":36,"rankRange":{"min":36,"max":36},"index":39.5,"indexRange":{"min":35.2,"max":41.9},"grade":"F","eligible":true,"verification":"independent","hardSetScore":31,"hardSetMean":31,"hardSetCount":6,"openWeights":false,"firstParty":false,"releaseDate":"2025-06-17","url":"https://opencharts.com/benchmarks/gemini-2-5-pro","categories":[{"category":"reasoning","mean":34.5,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":50.8,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":36.8,"benchmarkCount":3,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":35.7,"benchmarkCount":5,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":81.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":82.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8529040404040404,"display":"85.3%","normalized":85.2,"normalizedMeaning":"85.2% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FcqUZ6gcq2HBnShFRWvKjGr.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"arc-agi-2","value":0.0486,"display":"4.9%","normalized":5.1,"normalizedMeaning":"5.1% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"32k"},{"benchmark":"arc-agi","value":0.41,"display":"41.0%","normalized":41.6,"normalizedMeaning":"41.6% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"16k"},{"benchmark":"critpt","value":0.02,"display":"2.0%","normalized":6.2,"normalizedMeaning":"6.2% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"swe-bench-verified","value":0.5755693581780539,"display":"57.6%","normalized":69,"normalizedMeaning":"69.0% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FKL724uHFGbiXj4o5PHyLn2.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.428240740740741,"display":"42.8%","normalized":69,"normalizedMeaning":"69.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.5403,"display":"54.0%","normalized":58.2,"normalizedMeaning":"58.2% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"16k"},{"benchmark":"arena-webdev","value":1225.514082008788,"display":"1,226","normalized":7.2,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8416666666666667,"display":"84.2%","normalized":84.2,"normalizedMeaning":"84.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FiJp7PgYZa3PiXAV8UWd7MC.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.24561403508771928,"display":"24.6%","normalized":26.2,"normalizedMeaning":"26.2% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FFkywQ2KS4AkLPzjSPmhHcS.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath-tier-4","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2z7kL9DpfvY44T9Wo99FYx.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"terminal-bench","value":0.326,"display":"32.6%","normalized":38.5,"normalizedMeaning":"38.5% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"gdpval","value":0.233,"display":"23.3%","normalized":46.9,"normalizedMeaning":"46.9% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.066,"display":"6.6%","normalized":13.9,"normalizedMeaning":"13.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":573.6420000000003,"display":"$574","normalized":4.4,"normalizedMeaning":"4.4% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"deep-research-bench","value":0.415,"display":"41.5%","normalized":75,"normalizedMeaning":"75.0% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1246.4991668926318,"display":"1,246","normalized":81.1,"normalizedMeaning":"41% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1445.6212356081403,"display":"1,446","normalized":82.5,"normalizedMeaning":"41% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-5-pro","name":"GPT-5.5 Pro","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":93.9,"indexRange":{"min":92.5,"max":95.5},"grade":"A","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":87.5,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-23","url":"https://opencharts.com/benchmarks/gpt-5-5-pro","categories":[{"category":"reasoning","mean":93.9,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"math","mean":86.8,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"arc-agi-2","value":0.8458,"display":"84.6%","normalized":89,"normalizedMeaning":"89.0% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.965,"display":"96.5%","normalized":98,"normalizedMeaning":"98.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simplebench","value":0.7690000000000001,"display":"76.9%","normalized":93.9,"normalizedMeaning":"93.9% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lmcouncil.ai/benchmarks","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.305714285714286,"display":"30.6%","normalized":94.6,"normalizedMeaning":"94.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"frontiermath","value":0.8771929824561403,"display":"87.7%","normalized":93.6,"normalizedMeaning":"93.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath-tier-4","value":0.7804878048780488,"display":"78.0%","normalized":80,"normalizedMeaning":"80.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"}]},{"slug":"gpt-5-4-pro","name":"GPT-5.4 Pro","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":93.4,"indexRange":{"min":92.5,"max":94.6},"grade":"A","eligible":false,"verification":"independent","hardSetScore":82.8,"hardSetMean":82.8,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-05","url":"https://opencharts.com/benchmarks/gpt-5-4-pro","categories":[{"category":"reasoning","mean":93.4,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":61.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":61.8,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":74,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.946,"display":"94.6%","normalized":98.3,"normalizedMeaning":"98.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"hle","value":0.4432,"display":"44.3%","normalized":95.3,"normalizedMeaning":"95.3% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05"},{"benchmark":"arc-agi-2","value":0.8333,"display":"83.3%","normalized":87.7,"normalizedMeaning":"87.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"arc-agi","value":0.945,"display":"94.5%","normalized":95.9,"normalizedMeaning":"95.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"simplebench","value":0.741,"display":"74.1%","normalized":90.5,"normalizedMeaning":"90.5% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05"},{"benchmark":"critpt","value":0.3,"display":"30.0%","normalized":92.9,"normalizedMeaning":"92.9% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"simpleqa-verified","value":0.463,"display":"46.3%","normalized":61.2,"normalizedMeaning":"61.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FRfSJbuuuMkeVHQj7xrth4u.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"weirdml","value":0.5744,"display":"57.4%","normalized":61.8,"normalizedMeaning":"61.8% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-05 · none"},{"benchmark":"frontiermath","value":0.8245614035087719,"display":"82.5%","normalized":88,"normalizedMeaning":"88.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FZFVa8dw7oiir6PhcF8XZMo.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"},{"benchmark":"frontiermath-tier-4","value":0.5853658536585366,"display":"58.5%","normalized":60,"normalizedMeaning":"60.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FW6b3C8AeNC7CAFs5DQma8N.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-05 · xhigh"}]},{"slug":"ernie-5-1","name":"ERNIE 5.1","lab":"Baidu","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":88.8,"indexRange":{"min":88.8,"max":88.8},"grade":"A-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-05-01","url":"https://opencharts.com/benchmarks/ernie-5-1","categories":[{"category":"human-preference","mean":88.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arena-text","value":1468.0891739548538,"display":"1,468","normalized":88.8,"normalizedMeaning":"44% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gemini-3-deep-think","name":"Gemini 3 Deep Think","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":88.7,"indexRange":{"min":84.3,"max":93.3},"grade":"A-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":89,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-12","url":"https://opencharts.com/benchmarks/gemini-3-deep-think","categories":[{"category":"reasoning","mean":88.7,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true}],"scores":[{"benchmark":"arc-agi-2","value":0.8458,"display":"84.6%","normalized":89,"normalizedMeaning":"89.0% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arc-agi","value":0.96,"display":"96.0%","normalized":97.5,"normalizedMeaning":"97.5% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"critpt","value":0.257142857142857,"display":"25.7%","normalized":79.6,"normalizedMeaning":"79.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"}]},{"slug":"gpt-5-6-sol-pro","name":"GPT-5.6 Sol Pro","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":87.7,"indexRange":{"min":84.8,"max":89.4},"grade":"A-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":82.5,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-09","url":"https://opencharts.com/benchmarks/gpt-5-6-sol-pro","categories":[{"category":"reasoning","mean":87.5,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":96.3,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":82.5,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":84.4,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"simplebench","value":0.7170000000000001,"display":"71.7%","normalized":87.5,"normalizedMeaning":"87.5% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"proxhigh"},{"benchmark":"weirdml","value":0.8943000000000001,"display":"89.4%","normalized":96.3,"normalizedMeaning":"96.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"promax"},{"benchmark":"frontiermath-tier-4","value":0.8048780487804879,"display":"80.5%","normalized":82.5,"normalizedMeaning":"82.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FcdWwJSH3jVfrKyhUHmT54T.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"promax"},{"benchmark":"apex-agents","value":0.4,"display":"40.0%","normalized":84.4,"normalizedMeaning":"84.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"promax"}]},{"slug":"seed-2-0-pro","name":"Seed 2.0 Pro","lab":"ByteDance","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":84.7,"indexRange":{"min":84,"max":85.3},"grade":"B+","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-01","url":"https://opencharts.com/benchmarks/seed-2-0-pro","categories":[{"category":"multimodal","mean":84,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":85.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arena-vision","value":1256.8624788649763,"display":"1,257","normalized":84,"normalizedMeaning":"42% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1455.7950910387742,"display":"1,456","normalized":85.3,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-3-codex","name":"GPT-5.3 Codex","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":80.9,"indexRange":{"min":70.9,"max":80.9},"grade":"B+","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":88.8,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-05","url":"https://opencharts.com/benchmarks/gpt-5-3-codex","categories":[{"category":"coding","mean":64.8,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":80.9,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true}],"scores":[{"benchmark":"swe-bench-verified","value":0.7479338842975206,"display":"74.8%","normalized":89.6,"normalizedMeaning":"89.6% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F6w5buMbgeEqtRoE55cLZVv.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"weirdml","value":0.793,"display":"79.3%","normalized":85.4,"normalizedMeaning":"85.4% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1408.5703698392051,"display":"1,409","normalized":19.3,"normalizedMeaning":"10% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"codex-harness · board 2026-09-05"},{"benchmark":"terminal-bench","value":0.784,"display":"78.4%","normalized":92.6,"normalizedMeaning":"92.6% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: SageAgent"},{"benchmark":"metr-time-horizons","value":349.530732,"display":"5.8 h","normalized":84.2,"normalizedMeaning":"84.2% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.318,"display":"31.8%","normalized":67.1,"normalizedMeaning":"67.1% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":5940.124000000002,"display":"$5,940","normalized":79.6,"normalizedMeaning":"79.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"qwen3-6-max","name":"Qwen3.6 Max","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":76.3,"indexRange":{"min":70.9,"max":81.6},"grade":"B","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":91.8,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-20","url":"https://opencharts.com/benchmarks/qwen3-6-max","categories":[{"category":"reasoning","mean":82.5,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":68.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":59.7,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":91.1,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":68.9,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":86.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8737373737373737,"display":"87.4%","normalized":88.1,"normalizedMeaning":"88.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJv2hJvjpvA8xJW3Bzv7bfw.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-preview"},{"benchmark":"simplebench","value":0.63,"display":"63.0%","normalized":76.9,"normalizedMeaning":"76.9% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-preview"},{"benchmark":"simpleqa-verified","value":0.5200400801603207,"display":"52.0%","normalized":68.8,"normalizedMeaning":"68.8% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FiyPiR8GUuyNNA6Fssziq32.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-preview"},{"benchmark":"swe-bench-verified","value":0.7665289256198347,"display":"76.7%","normalized":91.8,"normalizedMeaning":"91.8% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FFkDMVLdEnJDR5PyUCT6BNQ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-preview"},{"benchmark":"arena-webdev","value":1478.7545108115191,"display":"1,479","normalized":27.6,"normalizedMeaning":"14% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-preview · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9111111111111112,"display":"91.1%","normalized":91.1,"normalizedMeaning":"91.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FEfJpx5HztMmL5EQSwsyqMe.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-preview"},{"benchmark":"vending-bench-2","value":4254.1866666666665,"display":"$4,254","normalized":68.9,"normalizedMeaning":"68.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-preview"},{"benchmark":"arena-text","value":1460.1325489829553,"display":"1,460","normalized":86.5,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-preview · board 2026-09-02"}]},{"slug":"muse-spark-1-2","name":"Muse Spark 1.2","lab":"Meta","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":75.2,"indexRange":{"min":70.7,"max":81.6},"grade":"B","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-08-05","url":"https://opencharts.com/benchmarks/muse-spark-1-2","categories":[{"category":"reasoning","mean":72.9,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":79.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":64,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":43,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":93.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":97.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"simplebench","value":0.745,"display":"74.5%","normalized":91,"normalizedMeaning":"91.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.177142857142857,"display":"17.7%","normalized":54.8,"normalizedMeaning":"54.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"simpleqa-verified","value":0.6030150753768844,"display":"60.3%","normalized":79.8,"normalizedMeaning":"79.8% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmLLmVVvo6t23MM5yxq7ryW.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"scicode","value":0.563657407407407,"display":"56.4%","normalized":90.9,"normalizedMeaning":"90.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"weirdml","value":0.603,"display":"60.3%","normalized":64.9,"normalizedMeaning":"64.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"arena-webdev","value":1534.1885008047357,"display":"1,534","normalized":36.1,"normalizedMeaning":"18% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-09-05"},{"benchmark":"proofbench","value":0.43,"display":"43.0%","normalized":43,"normalizedMeaning":"43.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1291.7629488182256,"display":"1,292","normalized":93.9,"normalizedMeaning":"47% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-08-27"},{"benchmark":"arena-text","value":1498.9914291515583,"display":"1,499","normalized":97.6,"normalizedMeaning":"49% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-09-02"}]},{"slug":"gpt-5-2-pro","name":"GPT-5.2 Pro","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":73,"indexRange":{"min":63.4,"max":73},"grade":"B-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":61,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2025-12-11","url":"https://opencharts.com/benchmarks/gpt-5-2-pro","categories":[{"category":"reasoning","mean":73,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"math","mean":63.1,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"arc-agi-2","value":0.5416,"display":"54.2%","normalized":57,"normalizedMeaning":"57.0% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · high"},{"benchmark":"arc-agi","value":0.905,"display":"90.5%","normalized":91.9,"normalizedMeaning":"91.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"simplebench","value":0.574,"display":"57.4%","normalized":70.1,"normalizedMeaning":"70.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-12-11"},{"benchmark":"frontiermath","value":0.74,"display":"74.0%","normalized":79,"normalizedMeaning":"79.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · xhigh"},{"benchmark":"frontiermath-tier-4","value":0.46,"display":"46.0%","normalized":47.1,"normalizedMeaning":"47.1% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-12-11 · xhigh"}]},{"slug":"gemini-3-pro","name":"Gemini 3 Pro","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":72.3,"indexRange":{"min":69.4,"max":76.9},"grade":"B-","eligible":false,"verification":"independent","hardSetScore":73.6,"hardSetMean":73.6,"hardSetCount":6,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-18","url":"https://opencharts.com/benchmarks/gemini-3-pro","categories":[{"category":"reasoning","mean":66.6,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":61.7,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":55.7,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":78,"benchmarkCount":6,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":93,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":93.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9261363636363636,"display":"92.6%","normalized":95.5,"normalizedMeaning":"95.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FCg54udQGcTCnfDGX2qePLZ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"hle","value":0.3752,"display":"37.5%","normalized":80.7,"normalizedMeaning":"80.7% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arc-agi-2","value":0.3111,"display":"31.1%","normalized":32.7,"normalizedMeaning":"32.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arc-agi","value":0.75,"display":"75.0%","normalized":76.1,"normalizedMeaning":"76.1% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"simplebench","value":0.7640000000000001,"display":"76.4%","normalized":93.3,"normalizedMeaning":"93.3% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"critpt","value":0.069,"display":"6.9%","normalized":21.4,"normalizedMeaning":"21.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"swe-bench-verified","value":0.7293388429752066,"display":"72.9%","normalized":87.4,"normalizedMeaning":"87.4% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F5XdYZBG5Wy8DUMRqcYv4fn.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"weirdml","value":0.6993,"display":"69.9%","normalized":75.3,"normalizedMeaning":"75.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"arena-webdev","value":1438.5512837238227,"display":"1,439","normalized":22.5,"normalizedMeaning":"11% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9138888888888888,"display":"91.4%","normalized":91.4,"normalizedMeaning":"91.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F4QM4h7ndQiDRL2anjKeddK.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"preview"},{"benchmark":"proofbench","value":0.2,"display":"20.0%","normalized":20,"normalizedMeaning":"20.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.6940000000000001,"display":"69.4%","normalized":81.9,"normalizedMeaning":"81.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · agent: Ante"},{"benchmark":"gdpval","value":0.403,"display":"40.3%","normalized":81.1,"normalizedMeaning":"81.1% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"metr-time-horizons","value":224.325884,"display":"3.7 h","normalized":77.9,"normalizedMeaning":"77.9% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/research","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"apex-agents","value":0.315,"display":"31.5%","normalized":66.5,"normalizedMeaning":"66.5% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"vending-bench-2","value":5478.158000000001,"display":"$5,478","normalized":77,"normalizedMeaning":"77.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview"},{"benchmark":"deep-research-bench","value":0.463,"display":"46.3%","normalized":83.7,"normalizedMeaning":"83.7% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · low"},{"benchmark":"arena-vision","value":1288.5878427882274,"display":"1,289","normalized":93,"normalizedMeaning":"47% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1485.576118119982,"display":"1,486","normalized":93.8,"normalizedMeaning":"47% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"muse-spark","name":"Muse Spark","lab":"Meta","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":71.3,"indexRange":{"min":71.3,"max":82.9},"grade":"B-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":87.2,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-08","url":"https://opencharts.com/benchmarks/muse-spark","categories":[{"category":"reasoning","mean":71.3,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":83,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":53,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":94.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":94.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.898,"display":"89.8%","normalized":91.6,"normalizedMeaning":"91.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"hle","value":0.4056,"display":"40.6%","normalized":87.2,"normalizedMeaning":"87.2% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.11330612244898002,"display":"11.3%","normalized":35.1,"normalizedMeaning":"35.1% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.515046296296296,"display":"51.5%","normalized":83,"normalizedMeaning":"83.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aime-2024-2025","value":0.889,"display":"88.9%","normalized":88.9,"normalizedMeaning":"88.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.17,"display":"17.0%","normalized":17,"normalizedMeaning":"17.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1293.7335999830548,"display":"1,294","normalized":94.4,"normalizedMeaning":"47% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1488.1683718271418,"display":"1,488","normalized":94.5,"normalizedMeaning":"47% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"muse-spark-1-1","name":"Muse Spark 1.1","lab":"Meta","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":71.3,"indexRange":{"min":67.3,"max":76.7},"grade":"B-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-09","url":"https://opencharts.com/benchmarks/muse-spark-1-1","categories":[{"category":"reasoning","mean":46.7,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":76.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":65.6,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":39,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":85.5,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":90.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":95.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.151,"display":"15.1%","normalized":46.7,"normalizedMeaning":"46.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.5778894472361809,"display":"57.8%","normalized":76.4,"normalizedMeaning":"76.4% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbCdwpRXgZtvQLYC4wA9JXr.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.582175925925926,"display":"58.2%","normalized":93.8,"normalizedMeaning":"93.8% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1541.0075339882792,"display":"1,541","normalized":37.3,"normalizedMeaning":"19% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.39,"display":"39.0%","normalized":39,"normalizedMeaning":"39.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning effort: xhigh"},{"benchmark":"apex-agents","value":0.419,"display":"41.9%","normalized":88.4,"normalizedMeaning":"88.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":6520.474999999999,"display":"$6,520","normalized":82.6,"normalizedMeaning":"82.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1279.4313032854543,"display":"1,279","normalized":90.4,"normalizedMeaning":"45% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1492.058078698562,"display":"1,492","normalized":95.7,"normalizedMeaning":"48% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-8-max","name":"Qwen3.8 Max","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":71.2,"indexRange":{"min":61.8,"max":79.1},"grade":"B-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":63.7,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2026-08-02","url":"https://opencharts.com/benchmarks/qwen3-8-max","categories":[{"category":"reasoning","mean":78.8,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":60.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":77.2,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":71.2,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"multimodal","mean":96.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":92.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9267676767676768,"display":"92.7%","normalized":95.6,"normalizedMeaning":"95.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max · xhigh"},{"benchmark":"critpt","value":0.2,"display":"20.0%","normalized":61.9,"normalizedMeaning":"61.9% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.458,"display":"45.8%","normalized":60.6,"normalizedMeaning":"60.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fbfftd5EzUK9yAekafEofvK.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max · xhigh"},{"benchmark":"scicode","value":0.528935185185185,"display":"52.9%","normalized":85.3,"normalizedMeaning":"85.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arena-webdev","value":1686.0702754308386,"display":"1,686","normalized":69.1,"normalizedMeaning":"35% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9944444444444444,"display":"99.4%","normalized":99.4,"normalizedMeaning":"99.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max · xhigh"},{"benchmark":"frontiermath","value":0.7473684210526316,"display":"74.7%","normalized":79.8,"normalizedMeaning":"79.8% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FABT89ieSWGnGj4RYMduHiG.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max · xhigh"},{"benchmark":"frontiermath-tier-4","value":0.4634146341463415,"display":"46.3%","normalized":47.5,"normalizedMeaning":"47.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FPBRdMZyz4L4CqPpXyLD2Lk.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max · xhigh"},{"benchmark":"proofbench","value":0.58,"display":"58.0%","normalized":58,"normalizedMeaning":"58.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arena-vision","value":1299.7116279072663,"display":"1,300","normalized":96.2,"normalizedMeaning":"48% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-08-27"},{"benchmark":"arena-text","value":1479.5567575891043,"display":"1,480","normalized":92.1,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-02"}]},{"slug":"gemini-3-8-flash","name":"Gemini 3.8 Flash","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":68.9,"indexRange":{"min":59.8,"max":75.8},"grade":"C+","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-09-02","url":"https://opencharts.com/benchmarks/gemini-3-8-flash","categories":[{"category":"reasoning","mean":56.6,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":74.7,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":48,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"human-preference","mean":96.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.182857142857143,"display":"18.3%","normalized":56.6,"normalizedMeaning":"56.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"scicode","value":0.543981481481482,"display":"54.4%","normalized":87.7,"normalizedMeaning":"87.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"medium"},{"benchmark":"cursorbench","value":0.6920000000000001,"display":"69.2%","normalized":94.3,"normalizedMeaning":"94.3% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · reasoning level: High"},{"benchmark":"arena-webdev","value":1567.230734798693,"display":"1,567","normalized":42.1,"normalizedMeaning":"21% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-05"},{"benchmark":"proofbench","value":0.48,"display":"48.0%","normalized":48,"normalizedMeaning":"48.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1493.847474085215,"display":"1,494","normalized":96.2,"normalizedMeaning":"48% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"qwen3-7-flash","name":"Qwen3.7 Flash","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":67.4,"indexRange":{"min":50.8,"max":83.9},"grade":"C+","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":20.6,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-27","url":"https://opencharts.com/benchmarks/qwen3-7-flash","categories":[{"category":"reasoning","mean":81,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"math","mean":53.7,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8232323232323232,"display":"82.3%","normalized":81,"normalizedMeaning":"81.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fi6i3gQf3ng2aAipvowPpws.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"aime-2024-2025","value":0.8666666666666667,"display":"86.7%","normalized":86.7,"normalizedMeaning":"86.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTjoS7sNuqVTc4LVyGRo4wa.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.19298245614035087,"display":"19.3%","normalized":20.6,"normalizedMeaning":"20.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FdoNrgGQY5SytQV22WRqw5L.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"}]},{"slug":"qwen3-5-397b-a17b","name":"Qwen3.5 397B-A17B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":65.7,"indexRange":{"min":60.2,"max":77.6},"grade":"C+","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":33.3,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-13","url":"https://opencharts.com/benchmarks/qwen3-5-397b-a17b","categories":[{"category":"reasoning","mean":86.7,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":18.3,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":61.1,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":81.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":81.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8636363636363636,"display":"86.4%","normalized":86.7,"normalizedMeaning":"86.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbXrmQqzF9sXbuZfRrSc7NP.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"arena-webdev","value":1398.55126177761,"display":"1,399","normalized":18.3,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8888888888888888,"display":"88.9%","normalized":88.9,"normalizedMeaning":"88.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJYuFdyZZSiMvmSZNzMTywb.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.312280701754386,"display":"31.2%","normalized":33.3,"normalizedMeaning":"33.3% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FA4yUA9PfzRr92JAqDDo9ro.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"arena-vision","value":1247.4081355604449,"display":"1,247","normalized":81.3,"normalizedMeaning":"41% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1441.105158531244,"display":"1,441","normalized":81.2,"normalizedMeaning":"41% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-7-max","name":"Qwen3.7 Max","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":65,"indexRange":{"min":56.4,"max":70.1},"grade":"C+","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":65.5,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-05-19","url":"https://opencharts.com/benchmarks/qwen3-7-max","categories":[{"category":"reasoning","mean":73.6,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":73.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":68.2,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":56.4,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"human-preference","mean":90.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9090909090909092,"display":"90.9%","normalized":93.1,"normalizedMeaning":"93.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FW6mTanyjvXhasFBzHukv4S.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"simplebench","value":0.7040000000000001,"display":"70.4%","normalized":86,"normalizedMeaning":"86.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"critpt","value":0.134285714285714,"display":"13.4%","normalized":41.6,"normalizedMeaning":"41.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.5576730190571715,"display":"55.8%","normalized":73.8,"normalizedMeaning":"73.8% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FDMjUrG6eZEqwcQ4SqUoViu.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"swe-bench-verified","value":0.7727272727272727,"display":"77.3%","normalized":92.6,"normalizedMeaning":"92.6% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F83hhDhcT9UUrZ2NiUbWSUU.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.488425925925926,"display":"48.8%","normalized":78.7,"normalizedMeaning":"78.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arena-webdev","value":1516.969927101015,"display":"1,517","normalized":33.3,"normalizedMeaning":"17% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-20260517 · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9555555555555556,"display":"95.6%","normalized":95.6,"normalizedMeaning":"95.6% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FD3YDUZSfEUDsS4frLbEKDu.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.6456140350877193,"display":"64.6%","normalized":68.9,"normalizedMeaning":"68.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FRG2biwnmGybSVTVxxMdRWa.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.34146341463414637,"display":"34.1%","normalized":35,"normalizedMeaning":"35.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FFHZCKbtfvpKGXU99vZFnXN.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.26,"display":"26.0%","normalized":26,"normalizedMeaning":"26.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arena-text","value":1473.618546639378,"display":"1,474","normalized":90.4,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-preview · board 2026-09-02"}]},{"slug":"qwen3-max","name":"Qwen3 Max","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":64.2,"indexRange":{"min":51.6,"max":64.2},"grade":"C","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":20.2,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2025-09-24","url":"https://opencharts.com/benchmarks/qwen3-max","categories":[{"category":"reasoning","mean":67.3,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":64.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"math","mean":64.2,"benchmarkCount":3,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":0,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":79.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.726010101010101,"display":"72.6%","normalized":67.3,"normalizedMeaning":"67.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F46RCssznDADYeg8MuSgvsj.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-2025-09-23"},{"benchmark":"simpleqa-verified","value":0.4874874874874875,"display":"48.7%","normalized":64.5,"normalizedMeaning":"64.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FfXG2GPt5F25tHRTPhAoZ4K.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-2025-09-23"},{"benchmark":"aime-2024-2025","value":0.7333333333333333,"display":"73.3%","normalized":73.3,"normalizedMeaning":"73.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FULeuPtwp3ZoyuXQYuuQ82i.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-2025-09-23"},{"benchmark":"math-level-5","value":0.9712990936555892,"display":"97.1%","normalized":99,"normalizedMeaning":"99.0% of the best (98.1%)","bestValue":0.9813066465256798,"fieldSize":6,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/math-level-5","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-2025-09-23"},{"benchmark":"frontiermath","value":0.18947368421052632,"display":"18.9%","normalized":20.2,"normalizedMeaning":"20.2% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FNrTH3EYHjp4hPdxgpCPGTX.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max-2025-09-23"},{"benchmark":"vending-bench-2","value":71.56500000000018,"display":"$72","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-2025-09-23"},{"benchmark":"arena-text","value":1434.62123202103,"display":"1,435","normalized":79.4,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max-preview · board 2026-09-02"}]},{"slug":"qwen3-6-plus","name":"Qwen3.6 Plus","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":64.1,"indexRange":{"min":57.4,"max":70.9},"grade":"C","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":55.1,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-31","url":"https://opencharts.com/benchmarks/qwen3-6-plus","categories":[{"category":"reasoning","mean":49.2,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":58.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":53.4,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":67.1,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":74.8,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":81.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8838383838383839,"display":"88.4%","normalized":89.6,"normalizedMeaning":"89.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Ff6ZdMtRpdfskGbrBCzrKSn.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.0285714285714286,"display":"2.9%","normalized":8.8,"normalizedMeaning":"8.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.44144144144144143,"display":"44.1%","normalized":58.4,"normalizedMeaning":"58.4% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F4LWNorP8Fy8V8doxirNFcH.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"swe-bench-verified","value":0.5785123966942148,"display":"57.9%","normalized":69.3,"normalizedMeaning":"69.3% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F8HV4HP9EM5KuLSwzifuLye.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.407407407407407,"display":"40.7%","normalized":65.7,"normalizedMeaning":"65.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1460.013889744188,"display":"1,460","normalized":25.1,"normalizedMeaning":"13% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9333333333333332,"display":"93.3%","normalized":93.3,"normalizedMeaning":"93.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjYiGAFfUQKJuRStPzU3PX3.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.3824561403508772,"display":"38.2%","normalized":40.8,"normalizedMeaning":"40.8% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjbyVWnEvvFVNNc8AEF6WcQ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"vending-bench-2","value":5114.872000000001,"display":"$5,115","normalized":74.8,"normalizedMeaning":"74.8% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1443.743758119101,"display":"1,444","normalized":81.9,"normalizedMeaning":"41% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"grok-4-20","name":"Grok 4.20","lab":"xAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":63.2,"indexRange":{"min":42.9,"max":68},"grade":"C","eligible":false,"verification":"independent","hardSetScore":50.4,"hardSetMean":50.4,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-17","url":"https://opencharts.com/benchmarks/grok-4-20","categories":[{"category":"reasoning","mean":83.5,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":39.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":36.2,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":42.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":69.8,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":83.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":90.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8933080808080808,"display":"89.3%","normalized":90.9,"normalizedMeaning":"90.9% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"reasoning"},{"benchmark":"arc-agi-2","value":0.6514,"display":"65.1%","normalized":68.6,"normalizedMeaning":"68.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.895,"display":"89.5%","normalized":90.9,"normalizedMeaning":"90.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.302,"display":"30.2%","normalized":39.9,"normalizedMeaning":"39.9% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FGVm7khGvtdnQKGfkgAHD8q.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"reasoning"},{"benchmark":"weirdml","value":0.5226,"display":"52.3%","normalized":56.3,"normalizedMeaning":"56.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1373.6636514818324,"display":"1,374","normalized":16.1,"normalizedMeaning":"8% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9222222222222224,"display":"92.2%","normalized":92.2,"normalizedMeaning":"92.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"reasoning"},{"benchmark":"frontiermath","value":0.4491228070175438,"display":"44.9%","normalized":47.9,"normalizedMeaning":"47.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FFLDt53waLTeshC7PK9uVy2.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"reasoning"},{"benchmark":"frontiermath-tier-4","value":0.17073170731707318,"display":"17.1%","normalized":17.5,"normalizedMeaning":"17.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FSX5FCWMk3HF83gGsP3WFA6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"reasoning"},{"benchmark":"proofbench","value":0.14,"display":"14.0%","normalized":14,"normalizedMeaning":"14.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning"},{"benchmark":"terminal-bench","value":0.573,"display":"57.3%","normalized":67.7,"normalizedMeaning":"67.7% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: grok-cli"},{"benchmark":"vending-bench-2","value":4662.846666666667,"display":"$4,663","normalized":71.9,"normalizedMeaning":"71.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1255.8493310461456,"display":"1,256","normalized":83.7,"normalizedMeaning":"42% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning · board 2026-08-27"},{"benchmark":"arena-text","value":1474.5768967799013,"display":"1,475","normalized":90.6,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"beta1 · board 2026-09-02"}]},{"slug":"gemma-4-26b-a4b","name":"Gemma 4 26B A4B","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":63.1,"indexRange":{"min":56.3,"max":70},"grade":"C","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-04-02","url":"https://opencharts.com/benchmarks/gemma-4-26b-a4b","categories":[{"category":"reasoning","mean":34.1,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":39.2,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":82.2,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":79.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":80.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7323232323232324,"display":"73.2%","normalized":68.2,"normalizedMeaning":"68.2% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FUsx9KaiXvHZU4qrzgc4sLg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"minimal"},{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.4004629629629631,"display":"40.0%","normalized":64.6,"normalizedMeaning":"64.6% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.3517,"display":"35.2%","normalized":37.9,"normalizedMeaning":"37.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1361.2023531016964,"display":"1,361","normalized":15,"normalizedMeaning":"8% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8222222222222222,"display":"82.2%","normalized":82.2,"normalizedMeaning":"82.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmstwSH7noJfvhGcd6Lz5kd.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"minimal"},{"benchmark":"arena-vision","value":1241.5370545485212,"display":"1,242","normalized":79.7,"normalizedMeaning":"40% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1438.4478977774456,"display":"1,438","normalized":80.5,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"glm-5-3","name":"GLM 5.3","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":60.9,"indexRange":{"min":50.8,"max":71.2},"grade":"C","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":51.7,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2026-08-14","url":"https://opencharts.com/benchmarks/glm-5-3","categories":[{"category":"reasoning","mean":76.2,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":54.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":74.3,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":60.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":89.9,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":92.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9090909090909092,"display":"90.9%","normalized":93.1,"normalizedMeaning":"93.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"critpt","value":0.191428571428571,"display":"19.1%","normalized":59.3,"normalizedMeaning":"59.3% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"simpleqa-verified","value":0.41,"display":"41.0%","normalized":54.2,"normalizedMeaning":"54.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTY8R8FV7zCwByEipkqyg2P.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"scicode","value":0.564814814814815,"display":"56.5%","normalized":91,"normalizedMeaning":"91.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"weirdml","value":0.7540000000000001,"display":"75.4%","normalized":81.2,"normalizedMeaning":"81.2% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arena-webdev","value":1608.9613186750753,"display":"1,609","normalized":50.6,"normalizedMeaning":"25% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9111111111111112,"display":"91.1%","normalized":91.1,"normalizedMeaning":"91.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.6877192982456141,"display":"68.8%","normalized":73.4,"normalizedMeaning":"73.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.2926829268292683,"display":"29.3%","normalized":30,"normalizedMeaning":"30.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.49,"display":"49.0%","normalized":49,"normalizedMeaning":"49.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"vending-bench-2","value":8163.609999999998,"display":"$8,164","normalized":89.9,"normalizedMeaning":"89.9% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1482.037220401277,"display":"1,482","normalized":92.8,"normalizedMeaning":"46% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · board 2026-09-02"}]},{"slug":"kimi-k2","name":"Kimi K2","lab":"Moonshot AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":60.6,"indexRange":{"min":55.9,"max":64.6},"grade":"C","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":49.8,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2025-11-06","url":"https://opencharts.com/benchmarks/kimi-k2","categories":[{"category":"reasoning","mean":83.7,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":41.8,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":83.1,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":36,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":78.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":40.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8421717171717171,"display":"84.2%","normalized":83.7,"normalizedMeaning":"83.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmuGecFpAg9wHd6gexWJ99L.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"aider-polyglot","value":59.1,"display":"59.1%","normalized":67.2,"normalizedMeaning":"67.2% of the best (88.0%)","bestValue":88,"fieldSize":5,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://aider.chat/docs/leaderboards/","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.4279,"display":"42.8%","normalized":46.1,"normalizedMeaning":"46.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking"},{"benchmark":"arena-webdev","value":1322.4366771820569,"display":"1,322","normalized":12.2,"normalizedMeaning":"6% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8305555555555556,"display":"83.1%","normalized":83.1,"normalizedMeaning":"83.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F8BM7dCHaUZgcYJwZWiBhiQ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"terminal-bench","value":0.35700000000000004,"display":"35.7%","normalized":42.1,"normalizedMeaning":"42.1% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · agent: Terminus 2"},{"benchmark":"metr-time-horizons","value":54.194148,"display":"54 min","normalized":57.4,"normalizedMeaning":"57.4% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking"},{"benchmark":"apex-agents","value":0.040999999999999995,"display":"4.1%","normalized":8.6,"normalizedMeaning":"8.6% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking"},{"benchmark":"arena-text","value":1430.1151267102225,"display":"1,430","normalized":78.2,"normalizedMeaning":"39% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.406,"display":"40.6%","normalized":40.6,"normalizedMeaning":"40.6% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"hunyuan-3","name":"Hunyuan 3","lab":"Tencent","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":58.8,"indexRange":{"min":32.4,"max":85.2},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-06-01","url":"https://opencharts.com/benchmarks/hunyuan-3","categories":[{"category":"coding","mean":32.4,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"human-preference","mean":85.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1511.6468605447617,"display":"1,512","normalized":32.4,"normalizedMeaning":"16% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"arena-text","value":1455.3862122530145,"display":"1,455","normalized":85.2,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-pro","name":"GPT-5 Pro","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":58.5,"indexRange":{"min":52.9,"max":71.5},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":41.7,"hardSetMean":41.7,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2025-10-06","url":"https://opencharts.com/benchmarks/gpt-5-pro","categories":[{"category":"reasoning","mean":58.5,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":65,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":39.8,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"hle","value":0.3164,"display":"31.6%","normalized":68,"normalizedMeaning":"68.0% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-10-06"},{"benchmark":"arc-agi-2","value":0.1833,"display":"18.3%","normalized":19.3,"normalizedMeaning":"19.3% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-10-06"},{"benchmark":"arc-agi","value":0.702,"display":"70.2%","normalized":71.3,"normalizedMeaning":"71.3% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-10-06 · high"},{"benchmark":"simplebench","value":0.616,"display":"61.6%","normalized":75.2,"normalizedMeaning":"75.2% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-10-06 · high"},{"benchmark":"weirdml","value":0.6039,"display":"60.4%","normalized":65,"normalizedMeaning":"65.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-10-06 · high"},{"benchmark":"frontiermath","value":0.5578947368421052,"display":"55.8%","normalized":59.6,"normalizedMeaning":"59.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTFDQUeRKbWiEodRQRrKdpy.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-10-06 · high"},{"benchmark":"frontiermath-tier-4","value":0.1951219512195122,"display":"19.5%","normalized":20,"normalizedMeaning":"20.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FHfDahF4ZXTN6bjfDFwcHTU.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-10-06 · high"}]},{"slug":"kimi-k2-6","name":"Kimi K2.6","lab":"Moonshot AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":58.5,"indexRange":{"min":50.7,"max":64.1},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":59.7,"hardSetCount":3,"openWeights":true,"firstParty":false,"releaseDate":"2026-04-20","url":"https://opencharts.com/benchmarks/kimi-k2-6","categories":[{"category":"reasoning","mean":58.9,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":46.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":67,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":49.9,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":45.2,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":85.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":86.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9078282828282828,"display":"90.8%","normalized":93,"normalizedMeaning":"93.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FLjgJoDosTLkiGVT59smA7q.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.08,"display":"8.0%","normalized":24.8,"normalizedMeaning":"24.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.349,"display":"34.9%","normalized":46.2,"normalizedMeaning":"46.2% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"swe-bench-verified","value":0.7665289256198347,"display":"76.7%","normalized":91.8,"normalizedMeaning":"91.8% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbUPtAK7spEwvmiC8pi6afc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.534722222222222,"display":"53.5%","normalized":86.2,"normalizedMeaning":"86.2% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.5586,"display":"55.9%","normalized":60.1,"normalizedMeaning":"60.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"cursorbench","value":0.476,"display":"47.6%","normalized":64.9,"normalizedMeaning":"64.9% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1508.7620479697612,"display":"1,509","normalized":32,"normalizedMeaning":"16% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9611111111111112,"display":"96.1%","normalized":96.1,"normalizedMeaning":"96.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FBTeRdhgt6TFBj6QkGX2j6P.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.5719298245614035,"display":"57.2%","normalized":61,"normalizedMeaning":"61.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FBxrNfyMffkVUq4n7U5eK3Q.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath-tier-4","value":0.2564102564102564,"display":"25.6%","normalized":26.3,"normalizedMeaning":"26.3% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FeVKA2eavdRAbSN6AnzyBxA.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.16,"display":"16.0%","normalized":16,"normalizedMeaning":"16.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"osworld-2","value":0.046,"display":"4.6%","normalized":14.6,"normalizedMeaning":"14.6% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.189,"display":"18.9%","normalized":39.9,"normalizedMeaning":"39.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":6204.570000000001,"display":"$6,205","normalized":81,"normalizedMeaning":"81.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1262.819270812544,"display":"1,263","normalized":85.6,"normalizedMeaning":"43% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1460.7332189060337,"display":"1,461","normalized":86.7,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-7-plus","name":"Qwen3.7 Plus","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":58.4,"indexRange":{"min":52.8,"max":68.3},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":36.7,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-06-02","url":"https://opencharts.com/benchmarks/qwen3-7-plus","categories":[{"category":"reasoning","mean":58.6,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":46.2,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":65,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":8.9,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":86.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":85.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8787878787878788,"display":"87.9%","normalized":88.8,"normalizedMeaning":"88.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FURoWjw9Htc6QPwABVcrybw.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.0914285714285714,"display":"9.1%","normalized":28.3,"normalizedMeaning":"28.3% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.454861111111111,"display":"45.5%","normalized":73.3,"normalizedMeaning":"73.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"frontiercode","value":0.102,"display":"10.2%","normalized":19.1,"normalizedMeaning":"19.1% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: mini-swe-agent · reasoning effort: none"},{"benchmark":"aime-2024-2025","value":0.9333333333333332,"display":"93.3%","normalized":93.3,"normalizedMeaning":"93.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJvdw6GAXiS6T79bkpbk6Ra.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.34385964912280703,"display":"34.4%","normalized":36.7,"normalizedMeaning":"36.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FTt8tBXHoCEYqGLePWXkM8u.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"osworld-2","value":0.027999999999999997,"display":"2.8%","normalized":8.9,"normalizedMeaning":"8.9% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1265.8389166995066,"display":"1,266","normalized":86.5,"normalizedMeaning":"43% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1455.2942306103268,"display":"1,455","normalized":85.2,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gemma-4-31b","name":"Gemma 4 31B","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":57,"indexRange":{"min":51.3,"max":65.6},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-04-02","url":"https://opencharts.com/benchmarks/gemma-4-31b","categories":[{"category":"reasoning","mean":38.1,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":13.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":47.2,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":73.3,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":85.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":84.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7575757575757576,"display":"75.8%","normalized":71.7,"normalizedMeaning":"71.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FP5L8qbXo42vTvbU3UvTNjk.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"minimal"},{"benchmark":"critpt","value":0.0142857142857143,"display":"1.4%","normalized":4.4,"normalizedMeaning":"4.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.104,"display":"10.4%","normalized":13.8,"normalizedMeaning":"13.8% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FD5cbXacPnJ8akq2jAtifEQ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.434027777777778,"display":"43.4%","normalized":70,"normalizedMeaning":"70.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.5226,"display":"52.3%","normalized":56.3,"normalizedMeaning":"56.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1363.065714830294,"display":"1,363","normalized":15.2,"normalizedMeaning":"8% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.7333333333333333,"display":"73.3%","normalized":73.3,"normalizedMeaning":"73.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjFJucVunfLgoQqoFvPVQWP.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"minimal"},{"benchmark":"arena-vision","value":1261.2775608059096,"display":"1,261","normalized":85.2,"normalizedMeaning":"43% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1451.281433176619,"display":"1,451","normalized":84.1,"normalizedMeaning":"42% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"grok-4","name":"Grok 4","lab":"xAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":56.3,"indexRange":{"min":51.9,"max":63.7},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":39.8,"hardSetMean":39.8,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2025-07-09","url":"https://opencharts.com/benchmarks/grok-4","categories":[{"category":"reasoning","mean":61.5,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":69.9,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":84,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":51,"benchmarkCount":6,"coveredCount":8,"indexed":true,"counted":true},{"category":"multimodal","mean":64.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":73,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":96.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.87,"display":"87.0%","normalized":87.6,"normalizedMeaning":"87.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"arc-agi-2","value":0.15975,"display":"16.0%","normalized":16.8,"normalizedMeaning":"16.8% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.6667000000000001,"display":"66.7%","normalized":67.7,"normalizedMeaning":"67.7% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simplebench","value":0.605,"display":"60.5%","normalized":73.9,"normalizedMeaning":"73.9% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aider-polyglot","value":79.6,"display":"79.6%","normalized":90.5,"normalizedMeaning":"90.5% of the best (88.0%)","bestValue":88,"fieldSize":5,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://aider.chat/docs/leaderboards/#polyglot-leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.4573,"display":"45.7%","normalized":49.2,"normalizedMeaning":"49.2% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aime-2024-2025","value":0.84,"display":"84.0%","normalized":84,"normalizedMeaning":"84.0% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"terminal-bench","value":0.272,"display":"27.2%","normalized":32.1,"normalizedMeaning":"32.1% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: OpenHands"},{"benchmark":"gdpval","value":0.211,"display":"21.1%","normalized":42.5,"normalizedMeaning":"42.5% of the best (49.7%)","bestValue":0.49700000000000005,"fieldSize":8,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gdpval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"metr-time-horizons","value":110.075251,"display":"110 min","normalized":67.6,"normalizedMeaning":"67.6% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.152,"display":"15.2%","normalized":32.1,"normalizedMeaning":"32.1% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"cybench","value":0.43,"display":"43.0%","normalized":46.2,"normalizedMeaning":"46.2% of the best (93.0%)","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://data.x.ai/2025-08-20-grok-4-model-card.pdf","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"}},{"benchmark":"deep-research-bench","value":0.473,"display":"47.3%","normalized":85.5,"normalizedMeaning":"85.5% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://drb.futuresearch.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: True"},{"benchmark":"arena-vision","value":1183.5981920062616,"display":"1,184","normalized":64.4,"normalizedMeaning":"32% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1410.984770417744,"display":"1,411","normalized":73,"normalizedMeaning":"37% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.969,"display":"96.9%","normalized":96.9,"normalizedMeaning":"96.9% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"qwen3-6-27b","name":"Qwen3.6 27B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":56.3,"indexRange":{"min":42.4,"max":70.1},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":37.5,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2026-04-22","url":"https://opencharts.com/benchmarks/qwen3-6-27b","categories":[{"category":"reasoning","mean":44.4,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":60.1,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":64.3,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8585858585858586,"display":"85.9%","normalized":86,"normalizedMeaning":"86.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbuLf8Ne29xmzMZBRQkvrQG.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.00857142857142857,"display":"0.9%","normalized":2.7,"normalizedMeaning":"2.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none"},{"benchmark":"scicode","value":0.372685185185185,"display":"37.3%","normalized":60.1,"normalizedMeaning":"60.1% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none"},{"benchmark":"aime-2024-2025","value":0.9111111111111112,"display":"91.1%","normalized":91.1,"normalizedMeaning":"91.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F452pB5w5ea6uy92e2q2NMv.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.3508771929824561,"display":"35.1%","normalized":37.5,"normalizedMeaning":"37.5% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FVonoEhAsGSJ2FHU33qnhui.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}}]},{"slug":"qwen3-5-plus","name":"Qwen3.5 Plus","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":54.8,"indexRange":{"min":44.2,"max":61.9},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-16","url":"https://opencharts.com/benchmarks/qwen3-5-plus","categories":[{"category":"reasoning","mean":84.6,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":33.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"math","mean":86.7,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":14.4,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8484848484848485,"display":"84.8%","normalized":84.6,"normalizedMeaning":"84.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmGXYSUNCMisvaRzhQp6g2L.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simpleqa-verified","value":0.2535070140280561,"display":"25.4%","normalized":33.5,"normalizedMeaning":"33.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjrawnM2wHFbVfJFJsap4kS.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"aime-2024-2025","value":0.8666666666666667,"display":"86.7%","normalized":86.7,"normalizedMeaning":"86.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FcFbntXuxUtfreULvt7vEt7.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"apex-agents","value":0.136,"display":"13.6%","normalized":28.7,"normalizedMeaning":"28.7% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":0.5440000000000417,"display":"$1","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"qwen3-8-flash-next","name":"Qwen3.8 Flash Next","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":54.5,"indexRange":{"min":54.5,"max":54.5},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-09-02","url":"https://opencharts.com/benchmarks/qwen3-8-flash-next","categories":[{"category":"coding","mean":54.5,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1626.458229538196,"display":"1,626","normalized":54.5,"normalizedMeaning":"27% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"}]},{"slug":"muse-spark-1-3","name":"Muse Spark 1.3","lab":"Meta","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":53.6,"indexRange":{"min":53.6,"max":53.6},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-09-05","url":"https://opencharts.com/benchmarks/muse-spark-1-3","categories":[{"category":"coding","mean":53.6,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1622.478323192865,"display":"1,622","normalized":53.6,"normalizedMeaning":"27% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh · board 2026-09-05"}]},{"slug":"minimax-m3","name":"MiniMax M3","lab":"MiniMax","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":53.5,"indexRange":{"min":52.1,"max":58.9},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-06-01","url":"https://opencharts.com/benchmarks/minimax-m3","categories":[{"category":"reasoning","mean":53.5,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":43.1,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":44.6,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":30.9,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":78.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":81.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9090909090909092,"display":"90.9%","normalized":93.1,"normalizedMeaning":"93.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FhDESmyGFrbZDomjff3rqzD.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simplebench","value":0.458,"display":"45.8%","normalized":55.9,"normalizedMeaning":"55.9% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0371428571428571,"display":"3.7%","normalized":11.5,"normalizedMeaning":"11.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.45370370370370405,"display":"45.4%","normalized":73.1,"normalizedMeaning":"73.1% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"frontiercode","value":0.147,"display":"14.7%","normalized":27.5,"normalizedMeaning":"27.5% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: mini-swe-agent · reasoning effort: none"},{"benchmark":"arena-webdev","value":1487.300425499026,"display":"1,487","normalized":28.8,"normalizedMeaning":"14% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.7111111111111111,"display":"71.1%","normalized":71.1,"normalizedMeaning":"71.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F79ZP6iwoBaF5t9a5VRqPnw.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.18,"display":"18.0%","normalized":18,"normalizedMeaning":"18.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"osworld-2","value":0.046,"display":"4.6%","normalized":14.6,"normalizedMeaning":"14.6% of the best (31.4%)","bestValue":0.3143,"fieldSize":9,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://osworld-v2.xlang.ai/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":2157.7716666666665,"display":"$2,158","normalized":47.1,"normalizedMeaning":"47.1% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1237.2262688971537,"display":"1,237","normalized":78.5,"normalizedMeaning":"39% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1442.9482267954468,"display":"1,443","normalized":81.7,"normalizedMeaning":"41% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-codex","name":"GPT-5 Codex","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":53,"indexRange":{"min":47.3,"max":55.5},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":52.3,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2025-09-15","url":"https://opencharts.com/benchmarks/gpt-5-codex","categories":[{"category":"coding","mean":58.7,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":47.3,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"weirdml","value":0.5453,"display":"54.5%","normalized":58.7,"normalizedMeaning":"58.7% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"terminal-bench","value":0.443,"display":"44.3%","normalized":52.3,"normalizedMeaning":"52.3% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Codex CLI"},{"benchmark":"apex-agents","value":0.201,"display":"20.1%","normalized":42.4,"normalizedMeaning":"42.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"}]},{"slug":"qwen3-5-35b-a3b","name":"Qwen3.5 35B-A3B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":52.2,"indexRange":{"min":42.1,"max":62.3},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-24","url":"https://opencharts.com/benchmarks/qwen3-5-35b-a3b","categories":[{"category":"reasoning","mean":42.2,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":27.7,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":70,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"human-preference","mean":68.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8345959595959596,"display":"83.5%","normalized":82.6,"normalizedMeaning":"82.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FRWjuCuVGvaRd4vDjJJ6aP2.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.00571428571428571,"display":"0.6%","normalized":1.8,"normalizedMeaning":"1.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none"},{"benchmark":"scicode","value":0.292824074074074,"display":"29.3%","normalized":47.2,"normalizedMeaning":"47.2% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none"},{"benchmark":"arena-webdev","value":1249.924529221946,"display":"1,250","normalized":8.2,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.7,"display":"70.0%","normalized":70,"normalizedMeaning":"70.0% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fh9aoDxyWEhybvARHpyuoco.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"arena-text","value":1395.1520727044651,"display":"1,395","normalized":68.8,"normalizedMeaning":"34% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"kimi-k2-5","name":"Kimi K2.5","lab":"Moonshot AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":52.1,"indexRange":{"min":48,"max":56.4},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":51.1,"hardSetMean":51.1,"hardSetCount":4,"openWeights":true,"firstParty":false,"releaseDate":"2026-01-27","url":"https://opencharts.com/benchmarks/kimi-k2-5","categories":[{"category":"reasoning","mean":47.7,"benchmarkCount":6,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":45.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":56.4,"benchmarkCount":5,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":92.2,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":36.5,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":81.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":83.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":78.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.876,"display":"87.6%","normalized":88.5,"normalizedMeaning":"88.5% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FFmqGUq87N4JcuSbo6Uu3U5.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"hle","value":0.24370000000000003,"display":"24.4%","normalized":52.4,"normalizedMeaning":"52.4% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi-2","value":0.1181,"display":"11.8%","normalized":12.4,"normalizedMeaning":"12.4% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.6533,"display":"65.3%","normalized":66.3,"normalizedMeaning":"66.3% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simplebench","value":0.468,"display":"46.8%","normalized":57.1,"normalizedMeaning":"57.1% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0314285714285714,"display":"3.1%","normalized":9.7,"normalizedMeaning":"9.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.343,"display":"34.3%","normalized":45.4,"normalizedMeaning":"45.4% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FRbXXg4JEbTSBnzw3CMSHnD.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"swe-bench-verified","value":0.737603305785124,"display":"73.8%","normalized":88.4,"normalizedMeaning":"88.4% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F5wVjUb3yn6bdjPSfPEMuiZ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.48958333333333304,"display":"49.0%","normalized":78.9,"normalizedMeaning":"78.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.456,"display":"45.6%","normalized":49.1,"normalizedMeaning":"49.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"cursorbench","value":0.319,"display":"31.9%","normalized":43.5,"normalizedMeaning":"43.5% of the best (73.4%)","bestValue":0.7340000000000001,"fieldSize":21,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cursor.com/cursorbench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1436.1888070984123,"display":"1,436","normalized":22.3,"normalizedMeaning":"11% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.922,"display":"92.2%","normalized":92.2,"normalizedMeaning":"92.2% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FmbfZvrSNoXSJqCyD4VKqLe.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"terminal-bench","value":0.4320000000000001,"display":"43.2%","normalized":51,"normalizedMeaning":"51.0% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"apex-agents","value":0.14400000000000002,"display":"14.4%","normalized":30.4,"normalizedMeaning":"30.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":1198.458,"display":"$1,198","normalized":28.1,"normalizedMeaning":"28.1% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1249.5745965887331,"display":"1,250","normalized":81.9,"normalizedMeaning":"41% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-08-27"},{"benchmark":"arena-text","value":1450.582684957285,"display":"1,451","normalized":83.9,"normalizedMeaning":"42% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.781,"display":"78.1%","normalized":78.1,"normalizedMeaning":"78.1% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"glm-5","name":"GLM 5","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":51.1,"indexRange":{"min":38.5,"max":66.4},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":51.1,"hardSetCount":3,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-11","url":"https://opencharts.com/benchmarks/glm-5","categories":[{"category":"reasoning","mean":51.1,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":53.5,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":80,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":56.1,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":85.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8781725888324873,"display":"87.8%","normalized":88.8,"normalizedMeaning":"88.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fn4hyz75SVpwPEisKVVjvNg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"arc-agi-2","value":0.0486,"display":"4.9%","normalized":5.1,"normalizedMeaning":"5.1% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.4467,"display":"44.7%","normalized":45.4,"normalizedMeaning":"45.4% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simplebench","value":0.532,"display":"53.2%","normalized":65,"normalizedMeaning":"65.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"swe-bench-verified","value":0.7207792207792207,"display":"72.1%","normalized":86.4,"normalizedMeaning":"86.4% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FhKkJEu4ueLiNWsTtrJ78PS.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"weirdml","value":0.4817,"display":"48.2%","normalized":51.9,"normalizedMeaning":"51.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1435.718981264792,"display":"1,436","normalized":22.2,"normalizedMeaning":"11% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8,"display":"80.0%","normalized":80,"normalizedMeaning":"80.0% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fccbx7KQY9bTD563vzBXwSg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"terminal-bench","value":0.524,"display":"52.4%","normalized":61.9,"normalizedMeaning":"61.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"apex-agents","value":0.172,"display":"17.2%","normalized":36.3,"normalizedMeaning":"36.3% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":4432.122000000001,"display":"$4,432","normalized":70.2,"normalizedMeaning":"70.2% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1457.772757077609,"display":"1,458","normalized":85.9,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-8-27b","name":"Qwen3.8 27B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":51,"indexRange":{"min":43.1,"max":59.7},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-08-14","url":"https://opencharts.com/benchmarks/qwen3-8-27b","categories":[{"category":"reasoning","mean":16.8,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":59.7,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":16,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":82.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":79.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.0542857142857143,"display":"5.4%","normalized":16.8,"normalizedMeaning":"16.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"scicode","value":0.446759259259259,"display":"44.7%","normalized":72,"normalizedMeaning":"72.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"arena-webdev","value":1593.7282930220022,"display":"1,594","normalized":47.4,"normalizedMeaning":"24% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.16,"display":"16.0%","normalized":16,"normalizedMeaning":"16.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1251.485307005685,"display":"1,251","normalized":82.5,"normalizedMeaning":"41% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1435.8926626599018,"display":"1,436","normalized":79.8,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"inkling-small","name":"Inkling Small","lab":"Thinking Machines","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":50.8,"indexRange":{"min":42.6,"max":56.6},"grade":"C-","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":36.4,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-15","url":"https://opencharts.com/benchmarks/inkling-small","categories":[{"category":"reasoning","mean":60.8,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":25.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":48.8,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":40.7,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"multimodal","mean":70.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":71.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.88510101010101,"display":"88.5%","normalized":89.7,"normalizedMeaning":"89.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"arc-agi-2","value":0.4013888888888889,"display":"40.1%","normalized":42.3,"normalizedMeaning":"42.3% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"arc-agi","value":0.84,"display":"84.0%","normalized":85.3,"normalizedMeaning":"85.3% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"xhigh"},{"benchmark":"critpt","value":0.08285714285714291,"display":"8.3%","normalized":25.7,"normalizedMeaning":"25.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.191,"display":"19.1%","normalized":25.3,"normalizedMeaning":"25.3% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FeJfbDKqeg9kNXFoxpeV7HG.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"scicode","value":0.487268518518519,"display":"48.7%","normalized":78.5,"normalizedMeaning":"78.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1404.9961681527584,"display":"1,405","normalized":19,"normalizedMeaning":"10% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9,"display":"90.0%","normalized":90,"normalizedMeaning":"90.0% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath","value":0.4631578947368421,"display":"46.3%","normalized":49.4,"normalizedMeaning":"49.4% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FB6bZbqickq8o2fQcgDmzMo.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"frontiermath-tier-4","value":0.17073170731707318,"display":"17.1%","normalized":17.5,"normalizedMeaning":"17.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FDVD8G8DHNyzkowBzK7JvMt.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"xhigh"},{"benchmark":"proofbench","value":0.06,"display":"6.0%","normalized":6,"normalizedMeaning":"6.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning effort: 0.99"},{"benchmark":"arena-vision","value":1206.6211781172724,"display":"1,207","normalized":70.3,"normalizedMeaning":"35% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1406.655281058057,"display":"1,407","normalized":71.9,"normalizedMeaning":"36% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"nemotron-3-ultra","name":"Nemotron 3 Ultra","lab":"NVIDIA","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":49.8,"indexRange":{"min":41.3,"max":58.2},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-06-04","url":"https://opencharts.com/benchmarks/nemotron-3-ultra","categories":[{"category":"reasoning","mean":47.5,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":55.6,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":44.4,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":24.3,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":77.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8535353535353535,"display":"85.4%","normalized":85.3,"normalizedMeaning":"85.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbpZy8tAbh7UV6A6U7XZtBg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.0314285714285714,"display":"3.1%","normalized":9.7,"normalizedMeaning":"9.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.399305555555556,"display":"39.9%","normalized":64.4,"normalizedMeaning":"64.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.43450000000000005,"display":"43.5%","normalized":46.8,"normalizedMeaning":"46.8% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aime-2024-2025","value":0.8666666666666667,"display":"86.7%","normalized":86.7,"normalizedMeaning":"86.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FWYTWr3Xk4qxymhMsm7vRQ2.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.02,"display":"2.0%","normalized":2,"normalizedMeaning":"2.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.115,"display":"11.5%","normalized":24.3,"normalizedMeaning":"24.3% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1426.1144071561305,"display":"1,426","normalized":77.1,"normalizedMeaning":"39% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"glm-4-7","name":"GLM 4.7","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":48.6,"indexRange":{"min":47.3,"max":53.8},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":39.4,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2025-12-22","url":"https://opencharts.com/benchmarks/glm-4-7","categories":[{"category":"reasoning","mean":48.6,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":42.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":47.5,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":44.7,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":36,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":81.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8333333333333334,"display":"83.3%","normalized":82.4,"normalizedMeaning":"82.4% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F6phXzd4aCbnBUSQmBsaywa.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simplebench","value":0.477,"display":"47.7%","normalized":58.2,"normalizedMeaning":"58.2% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0171020408163265,"display":"1.7%","normalized":5.3,"normalizedMeaning":"5.3% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.322,"display":"32.2%","normalized":42.6,"normalizedMeaning":"42.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FaYau6EG7CCaWVSyVq3vvQc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"scicode","value":0.451388888888889,"display":"45.1%","normalized":72.8,"normalizedMeaning":"72.8% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1434.497144651133,"display":"1,434","normalized":22.1,"normalizedMeaning":"11% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8333333333333334,"display":"83.3%","normalized":83.3,"normalizedMeaning":"83.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FGYzmrzMfeSjNjm2J6L9PeT.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.06,"display":"6.0%","normalized":6,"normalizedMeaning":"6.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"terminal-bench","value":0.334,"display":"33.4%","normalized":39.4,"normalizedMeaning":"39.4% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"apex-agents","value":0.087,"display":"8.7%","normalized":18.4,"normalizedMeaning":"18.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":2376.816,"display":"$2,377","normalized":50.2,"normalizedMeaning":"50.2% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1441.8072888362249,"display":"1,442","normalized":81.4,"normalizedMeaning":"41% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"deepseek-v3-1-terminus","name":"DeepSeek V3.1 Terminus","lab":"DeepSeek","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":48.5,"indexRange":{"min":35.4,"max":70.2},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2025-09-22","url":"https://opencharts.com/benchmarks/deepseek-v3-1-terminus","categories":[{"category":"reasoning","mean":5.3,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":65.5,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"human-preference","mean":74.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.0171428571428571,"display":"1.7%","normalized":5.3,"normalizedMeaning":"5.3% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.40625,"display":"40.6%","normalized":65.5,"normalizedMeaning":"65.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1417.527989564742,"display":"1,418","normalized":74.8,"normalizedMeaning":"37% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-09-02"}]},{"slug":"deepseek-v3-2","name":"DeepSeek V3.2","lab":"DeepSeek","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":48.3,"indexRange":{"min":43.5,"max":53.3},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":25.5,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2025-12-01","url":"https://opencharts.com/benchmarks/deepseek-v3-2","categories":[{"category":"reasoning","mean":43.5,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":53.1,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":47.9,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":28.3,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":76.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8342352092352093,"display":"83.4%","normalized":82.6,"normalizedMeaning":"82.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FCDdjbqiYYPtEBGoZLMPM3n.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"arc-agi-2","value":0.0403,"display":"4.0%","normalized":4.2,"normalizedMeaning":"4.2% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.57,"display":"57.0%","normalized":57.9,"normalizedMeaning":"57.9% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simplebench","value":0.526,"display":"52.6%","normalized":64.2,"normalizedMeaning":"64.2% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0285714285714286,"display":"2.9%","normalized":8.8,"normalizedMeaning":"8.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"exp · thinking"},{"benchmark":"aider-polyglot","value":74.2,"display":"74.2%","normalized":84.3,"normalizedMeaning":"84.3% of the best (88.0%)","bestValue":88,"fieldSize":5,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://aider.chat/docs/leaderboards/#polyglot-leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"exp · thinking"},{"benchmark":"scicode","value":0.388888888888889,"display":"38.9%","normalized":62.7,"normalizedMeaning":"62.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"exp · thinking"},{"benchmark":"weirdml","value":0.4673,"display":"46.7%","normalized":50.3,"normalizedMeaning":"50.3% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1360.2852472834757,"display":"1,360","normalized":15,"normalizedMeaning":"8% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.878174603174603,"display":"87.8%","normalized":87.8,"normalizedMeaning":"87.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F8szQqvTsGG8gJ2TiFReRn9.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"proofbench","value":0.08,"display":"8.0%","normalized":8,"normalizedMeaning":"8.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning effort: none"},{"benchmark":"terminal-bench","value":0.396,"display":"39.6%","normalized":46.8,"normalizedMeaning":"46.8% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"apex-agents","value":0.07,"display":"7.0%","normalized":14.8,"normalizedMeaning":"14.8% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":1034.004,"display":"$1,034","normalized":23.4,"normalizedMeaning":"23.4% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"exp"},{"benchmark":"arena-text","value":1425.2463639973778,"display":"1,425","normalized":76.8,"normalizedMeaning":"38% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"glm-5-3-flash","name":"GLM 5.3 Flash","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":48,"indexRange":{"min":32.7,"max":58.2},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":38.6,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2026-08-20","url":"https://opencharts.com/benchmarks/glm-5-3-flash","categories":[{"category":"reasoning","mean":69.9,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":62,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":48,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"multimodal","mean":88.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":90.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.9015151515151516,"display":"90.2%","normalized":92.1,"normalizedMeaning":"92.1% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"critpt","value":0.154285714285714,"display":"15.4%","normalized":47.8,"normalizedMeaning":"47.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.460648148148148,"display":"46.1%","normalized":74.3,"normalizedMeaning":"74.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1604.9672594041967,"display":"1,605","normalized":49.7,"normalizedMeaning":"25% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9388888888888888,"display":"93.9%","normalized":93.9,"normalizedMeaning":"93.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath","value":0.5578947368421052,"display":"55.8%","normalized":59.6,"normalizedMeaning":"59.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"frontiermath-tier-4","value":0.17073170731707318,"display":"17.1%","normalized":17.5,"normalizedMeaning":"17.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/frontiermath","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"max"},{"benchmark":"proofbench","value":0.21,"display":"21.0%","normalized":21,"normalizedMeaning":"21.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"arena-vision","value":1272.631985453219,"display":"1,273","normalized":88.4,"normalizedMeaning":"44% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1474.1588937337372,"display":"1,474","normalized":90.5,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-5-122b-a10b","name":"Qwen3.5 122B-A10B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":47.3,"indexRange":{"min":37.9,"max":62.2},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-24","url":"https://opencharts.com/benchmarks/qwen3-5-122b-a10b","categories":[{"category":"reasoning","mean":2.7,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":36.2,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"multimodal","mean":75.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":74.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.00857142857142857,"display":"0.9%","normalized":2.7,"normalizedMeaning":"2.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none"},{"benchmark":"scicode","value":0.356481481481482,"display":"35.6%","normalized":57.5,"normalizedMeaning":"57.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"none"},{"benchmark":"arena-webdev","value":1357.8392130253717,"display":"1,358","normalized":14.8,"normalizedMeaning":"7% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"arena-vision","value":1226.5491695633186,"display":"1,227","normalized":75.6,"normalizedMeaning":"38% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1417.1488357462335,"display":"1,417","normalized":74.7,"normalizedMeaning":"37% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"grok-4-fast","name":"Grok 4 Fast","lab":"xAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":47.1,"indexRange":{"min":40.1,"max":51.4},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":5.6,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2025-09-19","url":"https://opencharts.com/benchmarks/grok-4-fast","categories":[{"category":"reasoning","mean":27.4,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":25.6,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":32.3,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":75,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":75,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arc-agi-2","value":0.05280000000000001,"display":"5.3%","normalized":5.6,"normalizedMeaning":"5.6% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.485,"display":"48.5%","normalized":49.2,"normalizedMeaning":"49.2% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.4286,"display":"42.9%","normalized":46.1,"normalizedMeaning":"46.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1161.8760046134848,"display":"1,162","normalized":5,"normalizedMeaning":"3% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning · board 2026-09-05"},{"benchmark":"cybench","value":0.3,"display":"30.0%","normalized":32.3,"normalizedMeaning":"32.3% of the best (93.0%)","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://data.x.ai/2025-09-19-grok-4-fast-model-card.pdf","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"}},{"benchmark":"arena-text","value":1418.2588787859604,"display":"1,418","normalized":75,"normalizedMeaning":"38% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.75,"display":"75.0%","normalized":75,"normalizedMeaning":"75.0% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"mimo-v2-5","name":"MiMo V2.5","lab":"Xiaomi","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":46.1,"indexRange":{"min":37.8,"max":54.8},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-05-01","url":"https://opencharts.com/benchmarks/mimo-v2-5","categories":[{"category":"reasoning","mean":11.5,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":45.9,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":16,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":77.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":79.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.0371428571428571,"display":"3.7%","normalized":11.5,"normalizedMeaning":"11.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.430555555555556,"display":"43.1%","normalized":69.4,"normalizedMeaning":"69.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1437.7051381505491,"display":"1,438","normalized":22.4,"normalizedMeaning":"11% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.16,"display":"16.0%","normalized":16,"normalizedMeaning":"16.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1234.8526751000388,"display":"1,235","normalized":77.9,"normalizedMeaning":"39% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1433.8090039662272,"display":"1,434","normalized":79.2,"normalizedMeaning":"40% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-6-flash","name":"Qwen3.6 Flash","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":46,"indexRange":{"min":35.9,"max":58.5},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":24,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-27","url":"https://opencharts.com/benchmarks/qwen3-6-flash","categories":[{"category":"reasoning","mean":62.7,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":21.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"math","mean":54.2,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8333333333333334,"display":"83.3%","normalized":82.4,"normalizedMeaning":"82.4% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FH2FvRXagQRQmiNa7WrNMU2.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simplebench","value":0.352,"display":"35.2%","normalized":43,"normalizedMeaning":"43.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"simpleqa-verified","value":0.15915915915915915,"display":"15.9%","normalized":21.1,"normalizedMeaning":"21.1% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FFoMSS4FmH3VxpsvZiUJPCv.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"aime-2024-2025","value":0.8444444444444444,"display":"84.4%","normalized":84.4,"normalizedMeaning":"84.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F9zx2EXPTvPmxsL9qFCjP3x.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.2245614035087719,"display":"22.5%","normalized":24,"normalizedMeaning":"24.0% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FRoscWx4NUGVqcHdko6QhKd.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"}]},{"slug":"gemini-2-5-flash","name":"Gemini 2.5 Flash","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":45.2,"indexRange":{"min":38.4,"max":49.9},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":20.2,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2025-06-17","url":"https://opencharts.com/benchmarks/gemini-2-5-flash","categories":[{"category":"reasoning","mean":26.9,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":45.1,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":9,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":72.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":72.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"simplebench","value":0.412,"display":"41.2%","normalized":50.3,"normalizedMeaning":"50.3% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.011,"display":"1.1%","normalized":3.4,"normalizedMeaning":"3.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.4191,"display":"41.9%","normalized":45.1,"normalizedMeaning":"45.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"16k"},{"benchmark":"terminal-bench","value":0.171,"display":"17.1%","normalized":20.2,"normalizedMeaning":"20.2% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Mini-SWE-Agent"},{"benchmark":"apex-agents","value":0.018000000000000002,"display":"1.8%","normalized":3.8,"normalizedMeaning":"3.8% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":548.836,"display":"$549","normalized":3,"normalizedMeaning":"3.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1214.4942379310085,"display":"1,214","normalized":72.4,"normalizedMeaning":"36% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1409.7923898607776,"display":"1,410","normalized":72.7,"normalizedMeaning":"36% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"mimo-v2-5-pro","name":"MiMo V2.5 Pro","lab":"Xiaomi","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":44.3,"indexRange":{"min":29.5,"max":55},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-05-01","url":"https://opencharts.com/benchmarks/mimo-v2-5-pro","categories":[{"category":"reasoning","mean":12.4,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":54.1,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":22,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"human-preference","mean":88.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.04,"display":"4.0%","normalized":12.4,"normalizedMeaning":"12.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.502314814814815,"display":"50.2%","normalized":81,"normalizedMeaning":"81.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1474.895416095404,"display":"1,475","normalized":27.1,"normalizedMeaning":"14% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.22,"display":"22.0%","normalized":22,"normalizedMeaning":"22.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1468.10691960819,"display":"1,468","normalized":88.8,"normalizedMeaning":"44% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"mistral-large-3","name":"Mistral Large 3","lab":"Mistral AI","country":{"code":"FR","name":"France"},"rank":null,"rankRange":null,"index":44.1,"indexRange":{"min":34.3,"max":58.8},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2025-12-02","url":"https://opencharts.com/benchmarks/mistral-large-3","categories":[{"category":"reasoning","mean":0,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":32.9,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"multimodal","mean":69.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":73.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.362268518518519,"display":"36.2%","normalized":58.4,"normalizedMeaning":"58.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1229.2302833091712,"display":"1,229","normalized":7.3,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"arena-vision","value":1205.1236534948844,"display":"1,205","normalized":69.9,"normalizedMeaning":"35% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1413.6667924340622,"display":"1,414","normalized":73.7,"normalizedMeaning":"37% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-5-27b","name":"Qwen3.5 27B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":43.6,"indexRange":{"min":33.6,"max":58.1},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-24","url":"https://opencharts.com/benchmarks/qwen3-5-27b","categories":[{"category":"coding","mean":28.7,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":0,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":73.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":72.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"weirdml","value":0.3953,"display":"39.5%","normalized":42.6,"normalizedMeaning":"42.6% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1357.1338379876024,"display":"1,357","normalized":14.7,"normalizedMeaning":"7% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"vending-bench-2","value":201.9759999999999,"display":"$202","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1218.9119911392536,"display":"1,219","normalized":73.5,"normalizedMeaning":"37% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1407.9667666837065,"display":"1,408","normalized":72.2,"normalizedMeaning":"36% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"grok-4-1","name":"Grok 4.1","lab":"xAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":43.1,"indexRange":{"min":20.6,"max":61.3},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-17","url":"https://opencharts.com/benchmarks/grok-4-1","categories":[{"category":"coding","mean":6.6,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":34.5,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":88.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1210.6679340066728,"display":"1,211","normalized":6.6,"normalizedMeaning":"3% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-09-05"},{"benchmark":"apex-agents","value":0.128,"display":"12.8%","normalized":27,"normalizedMeaning":"27.0% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"cybench","value":0.39,"display":"39.0%","normalized":41.9,"normalizedMeaning":"41.9% of the best (93.0%)","bestValue":0.93,"fieldSize":7,"counted":true,"verification":"independent","provenance":"developer-reported","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://data.x.ai/2025-11-17-grok-4-1-model-card.pdf","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"developer-reported"}},{"benchmark":"arena-text","value":1465.4605316521504,"display":"1,465","normalized":88.1,"normalizedMeaning":"44% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"thinking · board 2026-09-02"}]},{"slug":"gpt-5-mini","name":"GPT-5 mini","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":43,"indexRange":{"min":37,"max":48.4},"grade":"D","eligible":false,"verification":"independent","hardSetScore":37.9,"hardSetMean":37.9,"hardSetCount":6,"openWeights":false,"firstParty":false,"releaseDate":"2025-08-07","url":"https://opencharts.com/benchmarks/gpt-5-mini","categories":[{"category":"reasoning","mean":34.5,"benchmarkCount":5,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":28.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":65.8,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":51.5,"benchmarkCount":5,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":20.6,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":64,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":67.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":62.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.75,"display":"75.0%","normalized":70.7,"normalizedMeaning":"70.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjP8EyLtPFsakgnsHe2oJM6.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"hle","value":0.1944,"display":"19.4%","normalized":41.8,"normalizedMeaning":"41.8% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07"},{"benchmark":"arc-agi-2","value":0.0444,"display":"4.4%","normalized":4.7,"normalizedMeaning":"4.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"arc-agi","value":0.5433,"display":"54.3%","normalized":55.2,"normalizedMeaning":"55.2% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07"},{"benchmark":"simpleqa-verified","value":0.216,"display":"21.6%","normalized":28.6,"normalizedMeaning":"28.6% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"swe-bench-verified","value":0.6466942148760331,"display":"64.7%","normalized":77.5,"normalizedMeaning":"77.5% of the best (83.5%)","bestValue":0.8347107438016529,"fieldSize":26,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F7nvdQy3AwKtoay2FBvfST5.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · medium"},{"benchmark":"scicode","value":0.392361111111111,"display":"39.2%","normalized":63.2,"normalizedMeaning":"63.2% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07"},{"benchmark":"weirdml","value":0.5267000000000001,"display":"52.7%","normalized":56.7,"normalizedMeaning":"56.7% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"aime-2024-2025","value":0.8666666666666667,"display":"86.7%","normalized":86.7,"normalizedMeaning":"86.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FR82gycJfm5jaCdh32tFkH4.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"math-level-5","value":0.978474320241692,"display":"97.8%","normalized":99.7,"normalizedMeaning":"99.7% of the best (98.1%)","bestValue":0.9813066465256798,"fieldSize":6,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/math-level-5","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"frontiermath","value":0.4666666666666667,"display":"46.7%","normalized":49.8,"normalizedMeaning":"49.8% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FYSR9WCR2SnJYvP7DJwX7YN.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"frontiermath-tier-4","value":0.12195121951219512,"display":"12.2%","normalized":12.5,"normalizedMeaning":"12.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2WxRbjq5ZWTgiXZPqYUG4R.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"proofbench","value":0.09,"display":"9.0%","normalized":9,"normalizedMeaning":"9.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.348,"display":"34.8%","normalized":41.1,"normalizedMeaning":"41.1% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · agent: spoox-m"},{"benchmark":"vending-bench-2","value":-31.18399999999995,"display":"$-31","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07"},{"benchmark":"arena-vision","value":1182.0783256260197,"display":"1,182","normalized":64,"normalizedMeaning":"32% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1389.4235936131686,"display":"1,389","normalized":67.4,"normalizedMeaning":"34% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.625,"display":"62.5%","normalized":62.5,"normalizedMeaning":"62.5% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · medium"}]},{"slug":"qwen3-6-35b-a3b","name":"Qwen3.6 35B-A3B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":42.9,"indexRange":{"min":32.4,"max":53.4},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":24.5,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2026-04-14","url":"https://opencharts.com/benchmarks/qwen3-6-35b-a3b","categories":[{"category":"reasoning","mean":42.8,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":47.4,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":54.2,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":27.2,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8484848484848485,"display":"84.8%","normalized":84.6,"normalizedMeaning":"84.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F9kErpmgQ42wTvgJaFn478v.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"critpt","value":0.00285714285714286,"display":"0.3%","normalized":0.9,"normalizedMeaning":"0.9% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.357638888888889,"display":"35.8%","normalized":57.6,"normalizedMeaning":"57.6% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.34490000000000004,"display":"34.5%","normalized":37.1,"normalizedMeaning":"37.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aime-2024-2025","value":0.8666666666666667,"display":"86.7%","normalized":86.7,"normalizedMeaning":"86.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F3n5KDcBaKqJh3WTg7qU2rs.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.20350877192982456,"display":"20.4%","normalized":21.7,"normalizedMeaning":"21.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FYRvn36dmMyDbS5e4XkF4Xi.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"terminal-bench","value":0.23,"display":"23.0%","normalized":27.2,"normalizedMeaning":"27.2% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: little-coder"}]},{"slug":"minimax-m2-1","name":"MiniMax M2.1","lab":"MiniMax","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":42.1,"indexRange":{"min":30.3,"max":54.6},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":43.2,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2025-12-23","url":"https://opencharts.com/benchmarks/minimax-m2-1","categories":[{"category":"coding","mean":17.3,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":43.2,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":65.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1387.274497440465,"display":"1,387","normalized":17.3,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-09-05"},{"benchmark":"terminal-bench","value":0.3660000000000001,"display":"36.6%","normalized":43.2,"normalizedMeaning":"43.2% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Crux"},{"benchmark":"arena-text","value":1383.8132122854472,"display":"1,384","normalized":65.9,"normalizedMeaning":"33% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-09-02"}]},{"slug":"grok-4-1-fast","name":"Grok 4.1 Fast","lab":"xAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":41.9,"indexRange":{"min":34.6,"max":49.4},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-19","url":"https://opencharts.com/benchmarks/grok-4-1-fast","categories":[{"category":"reasoning","mean":68.4,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":7.8,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":4,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":25.6,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":67.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":78.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"simplebench","value":0.56,"display":"56.0%","normalized":68.4,"normalizedMeaning":"68.4% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"non-reasoning"},{"benchmark":"arena-webdev","value":1240.179955504369,"display":"1,240","normalized":7.8,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning · board 2026-09-05"},{"benchmark":"proofbench","value":0.04,"display":"4.0%","normalized":4,"normalizedMeaning":"4.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning"},{"benchmark":"vending-bench-2","value":1106.6280000000002,"display":"$1,107","normalized":25.6,"normalizedMeaning":"25.6% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning"},{"benchmark":"arena-vision","value":1194.618457543359,"display":"1,195","normalized":67.2,"normalizedMeaning":"34% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning · board 2026-08-27"},{"benchmark":"arena-text","value":1430.2486980258063,"display":"1,430","normalized":78.2,"normalizedMeaning":"39% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning · board 2026-09-02"}]},{"slug":"grok-4-3","name":"Grok 4.3","lab":"xAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":41.3,"indexRange":{"min":23.9,"max":51.3},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":30.4,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-17","url":"https://opencharts.com/benchmarks/grok-4-3","categories":[{"category":"reasoning","mean":57.5,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":43.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":48.2,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":41.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":0,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":79.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":81.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8882575757575758,"display":"88.8%","normalized":90.2,"normalizedMeaning":"90.2% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"critpt","value":0.08,"display":"8.0%","normalized":24.8,"normalizedMeaning":"24.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"simpleqa-verified","value":0.332,"display":"33.2%","normalized":43.9,"normalizedMeaning":"43.9% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FY39ZRp553ULpUDBV2XrBbc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"scicode","value":0.47337962962963,"display":"47.3%","normalized":76.3,"normalizedMeaning":"76.3% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.4989,"display":"49.9%","normalized":53.7,"normalizedMeaning":"53.7% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1356.4933547551022,"display":"1,356","normalized":14.7,"normalizedMeaning":"7% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.9333333333333332,"display":"93.3%","normalized":93.3,"normalizedMeaning":"93.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.4280701754385965,"display":"42.8%","normalized":45.7,"normalizedMeaning":"45.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FbdsZjrG4ayzP8NAJg5q586.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath-tier-4","value":0.14634146341463414,"display":"14.6%","normalized":15,"normalizedMeaning":"15.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F3iRQkf5v7Wu7YuZhxa76Gg.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.11,"display":"11.0%","normalized":11,"normalizedMeaning":"11.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · reasoning effort: high"},{"benchmark":"vending-bench-2","value":35.26166666666665,"display":"$35","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-vision","value":1241.337655005879,"display":"1,241","normalized":79.6,"normalizedMeaning":"40% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1442.9637352784448,"display":"1,443","normalized":81.7,"normalizedMeaning":"41% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-2-codex","name":"GPT-5.2 Codex","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":40.9,"indexRange":{"min":35.8,"max":68.4},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":78.5,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2025-12-18","url":"https://opencharts.com/benchmarks/gpt-5-2-codex","categories":[{"category":"coding","mean":13.3,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":68.4,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1338.4921843348047,"display":"1,338","normalized":13.3,"normalizedMeaning":"7% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"terminal-bench","value":0.665,"display":"66.5%","normalized":78.5,"normalizedMeaning":"78.5% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Deep Agents"},{"benchmark":"apex-agents","value":0.276,"display":"27.6%","normalized":58.2,"normalizedMeaning":"58.2% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"gpt-5-1-codex-mini","name":"GPT-5.1 Codex Mini","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":40.3,"indexRange":{"min":7.9,"max":72.7},"grade":"D","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":72.7,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-13","url":"https://opencharts.com/benchmarks/gpt-5-1-codex-mini","categories":[{"category":"coding","mean":7.9,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":72.7,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1243.6765578759887,"display":"1,244","normalized":7.9,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"terminal-bench","value":0.616,"display":"61.6%","normalized":72.7,"normalizedMeaning":"72.7% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: hookele"}]},{"slug":"gpt-5-4-nano","name":"GPT-5.4 nano","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":39.5,"indexRange":{"min":31.2,"max":45.3},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":22.1,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-17","url":"https://opencharts.com/benchmarks/gpt-5-4-nano","categories":[{"category":"reasoning","mean":40.6,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":15.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":64.3,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":38.3,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":35.7,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":68.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":70.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7847,"display":"78.5%","normalized":75.6,"normalizedMeaning":"75.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FfZB3qGSohAVYzN97mniuUp.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · high"},{"benchmark":"arc-agi-2","value":0.0569,"display":"5.7%","normalized":6,"normalizedMeaning":"6.0% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"arc-agi","value":0.515,"display":"51.5%","normalized":52.3,"normalizedMeaning":"52.3% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"critpt","value":0.0925306122448979,"display":"9.3%","normalized":28.6,"normalizedMeaning":"28.6% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"simpleqa-verified","value":0.117,"display":"11.7%","normalized":15.5,"normalizedMeaning":"15.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F3QBsxKkbF2R2LEXTd6HDrJ.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · high"},{"benchmark":"scicode","value":0.46875,"display":"46.9%","normalized":75.6,"normalizedMeaning":"75.6% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · xhigh"},{"benchmark":"weirdml","value":0.4923,"display":"49.2%","normalized":53,"normalizedMeaning":"53.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · high"},{"benchmark":"aime-2024-2025","value":0.8778,"display":"87.8%","normalized":87.8,"normalizedMeaning":"87.8% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjVv9m8DjkdpZaLPdUzAnG9.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · high"},{"benchmark":"frontiermath","value":0.4491228070175438,"display":"44.9%","normalized":47.9,"normalizedMeaning":"47.9% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FiKtopjTjyCWZqJaEjcc7ZL.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · high"},{"benchmark":"frontiermath-tier-4","value":0.12195121951219512,"display":"12.2%","normalized":12.5,"normalizedMeaning":"12.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FSF6FYnnKztPCDWQc4v4EJa.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2026-03-17 · high"},{"benchmark":"proofbench","value":0.05,"display":"5.0%","normalized":5,"normalizedMeaning":"5.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17 · high · reasoning effort: high"},{"benchmark":"apex-agents","value":0.16899999999999998,"display":"16.9%","normalized":35.7,"normalizedMeaning":"35.7% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2026-03-17"},{"benchmark":"arena-vision","value":1201.453567646536,"display":"1,201","normalized":68.9,"normalizedMeaning":"34% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1402.2118864656431,"display":"1,402","normalized":70.7,"normalizedMeaning":"35% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"}]},{"slug":"qwen3-5-flash","name":"Qwen3.5 Flash","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":39.5,"indexRange":{"min":31.2,"max":47.4},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":19.5,"hardSetCount":1,"openWeights":false,"firstParty":false,"releaseDate":"2026-02-25","url":"https://opencharts.com/benchmarks/qwen3-5-flash","categories":[{"category":"reasoning","mean":81,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"knowledge","mean":26.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":7.7,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":52,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":0,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":69.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8232323232323232,"display":"82.3%","normalized":81,"normalizedMeaning":"81.0% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FS7wbot7JR53BAjM8F9gaws.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"simpleqa-verified","value":0.2032032032032032,"display":"20.3%","normalized":26.9,"normalizedMeaning":"26.9% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FK8zTckSXLKLesoLBv24TWc.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"arena-webdev","value":1237.962512812382,"display":"1,238","normalized":7.7,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8444444444444444,"display":"84.4%","normalized":84.4,"normalizedMeaning":"84.4% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FawDKyKbPXRzBVZzfdQucef.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.1824561403508772,"display":"18.2%","normalized":19.5,"normalizedMeaning":"19.5% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FafxoicJc4C6caVDwUpsiME.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"vending-bench-2","value":462.68500000000023,"display":"$463","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1396.9027397761977,"display":"1,397","normalized":69.3,"normalizedMeaning":"35% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-oss-20b","name":"GPT-OSS 20B","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":39.4,"indexRange":{"min":32.9,"max":48.2},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":4,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2025-08-05","url":"https://opencharts.com/benchmarks/gpt-oss-20b","categories":[{"category":"reasoning","mean":27.5,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":49.8,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":65.3,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":4,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":50.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.6079545454545454,"display":"60.8%","normalized":50.6,"normalizedMeaning":"50.6% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FGEwyLqZT93Hp55fdtHrJhm.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"medium"},{"benchmark":"critpt","value":0.0142857142857143,"display":"1.4%","normalized":4.4,"normalizedMeaning":"4.4% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"scicode","value":0.34375,"display":"34.4%","normalized":55.4,"normalizedMeaning":"55.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.4093,"display":"40.9%","normalized":44.1,"normalizedMeaning":"44.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"aime-2024-2025","value":0.6527777777777778,"display":"65.3%","normalized":65.3,"normalizedMeaning":"65.3% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FDB8ja6pEZWGocQbdciyisL.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"medium"},{"benchmark":"terminal-bench","value":0.034,"display":"3.4%","normalized":4,"normalizedMeaning":"4.0% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Mini-SWE-Agent"},{"benchmark":"arena-text","value":1317.1232025232875,"display":"1,317","normalized":50.2,"normalizedMeaning":"25% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"qwen3-5-9b","name":"Qwen3.5 9B","lab":"Alibaba","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":38.9,"indexRange":{"min":29.5,"max":48.3},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":10.9,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-24","url":"https://opencharts.com/benchmarks/qwen3-5-9b","categories":[{"category":"reasoning","mean":38.6,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":44.4,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":61.7,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":10.9,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7897727272727273,"display":"79.0%","normalized":76.3,"normalizedMeaning":"76.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fo6migB5WP27uLr9nUWpCiM.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0.00285714285714286,"display":"0.3%","normalized":0.9,"normalizedMeaning":"0.9% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.275462962962963,"display":"27.5%","normalized":44.4,"normalizedMeaning":"44.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aime-2024-2025","value":0.6166666666666667,"display":"61.7%","normalized":61.7,"normalizedMeaning":"61.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FYBMTDnW5LmCN6tBp2QYHzi.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"none"},{"benchmark":"terminal-bench","value":0.092,"display":"9.2%","normalized":10.9,"normalizedMeaning":"10.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: little-coder"}]},{"slug":"gpt-5-nano","name":"GPT-5 nano","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":35.8,"indexRange":{"min":29,"max":42.8},"grade":"F","eligible":false,"verification":"independent","hardSetScore":13.1,"hardSetMean":13.1,"hardSetCount":4,"openWeights":false,"firstParty":false,"releaseDate":"2025-08-07","url":"https://opencharts.com/benchmarks/gpt-5-nano","categories":[{"category":"reasoning","mean":28.8,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":15.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":41,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":42.8,"benchmarkCount":5,"coveredCount":5,"indexed":true,"counted":true},{"category":"agentic","mean":25.7,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":55.1,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":54.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"long-context","mean":21.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.6944444444444444,"display":"69.4%","normalized":62.8,"normalizedMeaning":"62.8% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2pSDjMSGDh4HsAxkqUuNQP.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"arc-agi-2","value":0.026100000000000005,"display":"2.6%","normalized":2.7,"normalizedMeaning":"2.7% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"arc-agi","value":0.2071,"display":"20.7%","normalized":21,"normalizedMeaning":"21.0% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · medium"},{"benchmark":"simpleqa-verified","value":0.117,"display":"11.7%","normalized":15.5,"normalizedMeaning":"15.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"weirdml","value":0.3806000000000001,"display":"38.1%","normalized":41,"normalizedMeaning":"41.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high"},{"benchmark":"aime-2024-2025","value":0.8111111111111111,"display":"81.1%","normalized":81.1,"normalizedMeaning":"81.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FY9YnaJQWpPFAPhe6jtKMhH.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"math-level-5","value":0.952416918429003,"display":"95.2%","normalized":97.1,"normalizedMeaning":"97.1% of the best (98.1%)","bestValue":0.9813066465256798,"fieldSize":6,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/math-level-5","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · medium"},{"benchmark":"frontiermath","value":0.2,"display":"20.0%","normalized":21.3,"normalizedMeaning":"21.3% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FJH3PjDrbrQndEcc8fh5Qre.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"frontiermath-tier-4","value":0.024390243902439025,"display":"2.4%","normalized":2.5,"normalizedMeaning":"2.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FP5v3FUnVVTbaqaCr43deKi.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"2025-08-07 · high"},{"benchmark":"proofbench","value":0.12,"display":"12.0%","normalized":12,"normalizedMeaning":"12.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · high · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.218,"display":"21.8%","normalized":25.7,"normalizedMeaning":"25.7% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · agent: spoox-o-m"},{"benchmark":"arena-vision","value":1145.214586968498,"display":"1,145","normalized":55.1,"normalizedMeaning":"28% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-08-27"},{"benchmark":"arena-text","value":1337.2183087441108,"display":"1,337","normalized":54.6,"normalizedMeaning":"27% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high · board 2026-09-02"},{"benchmark":"fiction-livebench","value":0.219,"display":"21.9%","normalized":21.9,"normalizedMeaning":"21.9% of the best (100.0%)","bestValue":1,"fieldSize":7,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://fiction.live/stories/Fiction-liveBench-Mar-14-2025/oQdzQvKHw8JyXbN87/home","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"2025-08-07 · medium"}]},{"slug":"minimax-m2-7","name":"MiniMax M2.7","lab":"MiniMax","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":35.4,"indexRange":{"min":25.7,"max":43.8},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":53.2,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2026-03-18","url":"https://opencharts.com/benchmarks/minimax-m2-7","categories":[{"category":"reasoning","mean":1.8,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":44.6,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":3,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":53.2,"benchmarkCount":1,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":74.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.00571428571428571,"display":"0.6%","normalized":1.8,"normalizedMeaning":"1.8% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.469907407407407,"display":"47.0%","normalized":75.7,"normalizedMeaning":"75.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.3695000000000001,"display":"37.0%","normalized":39.8,"normalizedMeaning":"39.8% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1398.4674589301735,"display":"1,398","normalized":18.3,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.03,"display":"3.0%","normalized":3,"normalizedMeaning":"3.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"terminal-bench","value":0.451,"display":"45.1%","normalized":53.2,"normalizedMeaning":"53.2% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: IndusAGI Coding Agent"},{"benchmark":"arena-text","value":1415.3621278943174,"display":"1,415","normalized":74.2,"normalizedMeaning":"37% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"glm-4-6","name":"GLM 4.6","lab":"Zhipu AI","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":34.2,"indexRange":{"min":20,"max":44.4},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":28.9,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2025-09-30","url":"https://opencharts.com/benchmarks/glm-4-6","categories":[{"category":"reasoning","mean":3.5,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":37.7,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":18.7,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":76.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0.0114285714285714,"display":"1.1%","normalized":3.5,"normalizedMeaning":"3.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.384259259259259,"display":"38.4%","normalized":61.9,"normalizedMeaning":"61.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1340.461324374597,"display":"1,340","normalized":13.5,"normalizedMeaning":"7% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"terminal-bench","value":0.245,"display":"24.5%","normalized":28.9,"normalizedMeaning":"28.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"apex-agents","value":0.04,"display":"4.0%","normalized":8.4,"normalizedMeaning":"8.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1424.8440332208606,"display":"1,425","normalized":76.7,"normalizedMeaning":"38% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gemini-3-1-flash-lite","name":"Gemini 3.1 Flash-Lite","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":34.1,"indexRange":{"min":34.1,"max":58.8},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":24.1,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2026-03-03","url":"https://opencharts.com/benchmarks/gemini-3-1-flash-lite","categories":[{"category":"reasoning","mean":34.1,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":44,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":54.8,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":47.5,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"multimodal","mean":77.9,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":78.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8181818181818182,"display":"81.8%","normalized":80.3,"normalizedMeaning":"80.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FnE4hmJo3QV6X5jeGWAeAze.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"hle","value":0.0864,"display":"8.6%","normalized":18.6,"normalizedMeaning":"18.6% of the best (46.5%)","bestValue":0.465,"fieldSize":18,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://lastexam.ai","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0114285714285714,"display":"1.1%","normalized":3.5,"normalizedMeaning":"3.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.418981481481481,"display":"41.9%","normalized":67.5,"normalizedMeaning":"67.5% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.5219,"display":"52.2%","normalized":56.2,"normalizedMeaning":"56.2% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1254.1790071341557,"display":"1,254","normalized":8.4,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.8,"display":"80.0%","normalized":80,"normalizedMeaning":"80.0% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FNMdaAXyqqPRmWfuxZUfHFF.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.2771929824561403,"display":"27.7%","normalized":29.6,"normalizedMeaning":"29.6% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FMVZf59jt9uEoWG9w5W9apB.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"apex-agents","value":0.13,"display":"13.0%","normalized":27.4,"normalizedMeaning":"27.4% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"deep-research-bench","value":0.373,"display":"37.3%","normalized":67.5,"normalizedMeaning":"67.5% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"low"},{"benchmark":"arena-vision","value":1235.0520876870658,"display":"1,235","normalized":77.9,"normalizedMeaning":"39% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-08-27"},{"benchmark":"arena-text","value":1432.4892570827278,"display":"1,432","normalized":78.8,"normalizedMeaning":"39% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"preview · board 2026-09-02"}]},{"slug":"mistral-medium-3-5","name":"Mistral Medium 3.5","lab":"Mistral AI","country":{"code":"FR","name":"France"},"rank":null,"rankRange":null,"index":33.7,"indexRange":{"min":33.7,"max":39.3},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":false,"releaseDate":"2026-04-28","url":"https://opencharts.com/benchmarks/mistral-medium-3-5","categories":[{"category":"reasoning","mean":0,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":33.7,"benchmarkCount":4,"coveredCount":7,"indexed":true,"counted":true},{"category":"math","mean":9,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"multimodal","mean":68.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":77.3,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.39583333333333304,"display":"39.6%","normalized":63.8,"normalizedMeaning":"63.8% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.4371,"display":"43.7%","normalized":47.1,"normalizedMeaning":"47.1% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"frontiercode","value":0.08,"display":"8.0%","normalized":15,"normalizedMeaning":"15.0% of the best (53.5%)","bestValue":0.5348,"fieldSize":27,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://cognition.com/frontiercode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"harness: chisel · reasoning effort: none"},{"benchmark":"arena-webdev","value":1265.0296802752273,"display":"1,265","normalized":8.9,"normalizedMeaning":"4% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.09,"display":"9.0%","normalized":9,"normalizedMeaning":"9.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning effort: high"},{"benchmark":"arena-vision","value":1199.037094545498,"display":"1,199","normalized":68.3,"normalizedMeaning":"34% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1426.8703335516295,"display":"1,427","normalized":77.3,"normalizedMeaning":"39% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-5-instant","name":"GPT-5.5 Instant","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":32.9,"indexRange":{"min":32.9,"max":69.5},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":15.3,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2026-05-05","url":"https://opencharts.com/benchmarks/gpt-5-5-instant","categories":[{"category":"reasoning","mean":40.7,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":78.4,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":32.9,"benchmarkCount":3,"coveredCount":5,"indexed":true,"counted":true},{"category":"multimodal","mean":89.8,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":90.4,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8251262626262627,"display":"82.5%","normalized":81.3,"normalizedMeaning":"81.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/gpqa-diamond","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.486111111111111,"display":"48.6%","normalized":78.4,"normalizedMeaning":"78.4% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"aime-2024-2025","value":0.6805555555555556,"display":"68.1%","normalized":68.1,"normalizedMeaning":"68.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/otis-mock-aime-2024-2025","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath","value":0.2631578947368421,"display":"26.3%","normalized":28.1,"normalizedMeaning":"28.1% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FgGcHjE6ppy54ahN2pfxsGH.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"frontiermath-tier-4","value":0.024390243902439025,"display":"2.4%","normalized":2.5,"normalizedMeaning":"2.5% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F2UmuBfhBvYbbnTg5BnQqov.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"}},{"benchmark":"arena-vision","value":1277.6426691199665,"display":"1,278","normalized":89.8,"normalizedMeaning":"45% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1473.7137198175467,"display":"1,474","normalized":90.4,"normalizedMeaning":"45% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"claude-haiku-4-5","name":"Claude Haiku 4.5","lab":"Anthropic","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":32.7,"indexRange":{"min":26.7,"max":37.6},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":23.1,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2025-10-15","url":"https://opencharts.com/benchmarks/claude-haiku-4-5","categories":[{"category":"reasoning","mean":29.5,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"knowledge","mean":17.5,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"coding","mean":43.8,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":82.5,"benchmarkCount":2,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":35.8,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true},{"category":"human-preference","mean":73.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7121212121212122,"display":"71.2%","normalized":65.3,"normalizedMeaning":"65.3% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FGEaAdb9UmoUFbEtBx7WzAR.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251001 · 32k"},{"benchmark":"arc-agi-2","value":0.0403,"display":"4.0%","normalized":4.2,"normalizedMeaning":"4.2% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001 · 32k"},{"benchmark":"arc-agi","value":0.4767,"display":"47.7%","normalized":48.4,"normalizedMeaning":"48.4% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001 · 32k"},{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001"},{"benchmark":"simpleqa-verified","value":0.132,"display":"13.2%","normalized":17.5,"normalizedMeaning":"17.5% of the best (75.6%)","bestValue":0.756,"fieldSize":57,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/simpleqa-verified","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251001"},{"benchmark":"scicode","value":0.43287037037037,"display":"43.3%","normalized":69.8,"normalizedMeaning":"69.8% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001"},{"benchmark":"weirdml","value":0.454,"display":"45.4%","normalized":48.9,"normalizedMeaning":"48.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001"},{"benchmark":"arena-webdev","value":1328.9038613398989,"display":"1,329","normalized":12.7,"normalizedMeaning":"6% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001 · board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.6666666666666666,"display":"66.7%","normalized":66.7,"normalizedMeaning":"66.7% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FneZvEzLi6yeL9z5KjGef7H.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251001 · 32k"},{"benchmark":"math-level-5","value":0.9636087945413192,"display":"96.4%","normalized":98.2,"normalizedMeaning":"98.2% of the best (98.1%)","bestValue":0.9813066465256798,"fieldSize":6,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/math-level-5","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"20251001 · 32k"},{"benchmark":"terminal-bench","value":0.355,"display":"35.5%","normalized":41.9,"normalizedMeaning":"41.9% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001 · agent: Goose"},{"benchmark":"apex-agents","value":0.08900000000000001,"display":"8.9%","normalized":18.8,"normalizedMeaning":"18.8% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001"},{"benchmark":"vending-bench-2","value":458.88599999999985,"display":"$459","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001"},{"benchmark":"deep-research-bench","value":0.455,"display":"45.5%","normalized":82.3,"normalizedMeaning":"82.3% of the best (55.3%)","bestValue":0.5529999999999999,"fieldSize":19,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://deepresearch-bench.github.io","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001 · low"},{"benchmark":"arena-text","value":1413.1776186661616,"display":"1,413","normalized":73.6,"normalizedMeaning":"37% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"20251001 · board 2026-09-02"}]},{"slug":"gemini-3-5-flash-lite","name":"Gemini 3.5 Flash-Lite","lab":"Google","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":32.5,"indexRange":{"min":24.9,"max":38.6},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":12.8,"hardSetCount":3,"openWeights":false,"firstParty":false,"releaseDate":"2026-07-21","url":"https://opencharts.com/benchmarks/gemini-3-5-flash-lite","categories":[{"category":"reasoning","mean":36.9,"benchmarkCount":4,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":43.9,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":28,"benchmarkCount":4,"coveredCount":5,"indexed":true,"counted":true},{"category":"multimodal","mean":86.7,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false},{"category":"human-preference","mean":85.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.8333333333333334,"display":"83.3%","normalized":82.4,"normalizedMeaning":"82.4% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fchbh6XstiqtUa5W9zuyfjS.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"arc-agi-2","value":0.1027777777777778,"display":"10.3%","normalized":10.8,"normalizedMeaning":"10.8% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arc-agi","value":0.535,"display":"53.5%","normalized":54.3,"normalizedMeaning":"54.3% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"critpt","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.40856481481481505,"display":"40.9%","normalized":65.9,"normalizedMeaning":"65.9% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.39,"display":"39.0%","normalized":42,"normalizedMeaning":"42.0% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"arena-webdev","value":1449.0146361191323,"display":"1,449","normalized":23.8,"normalizedMeaning":"12% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"aime-2024-2025","value":0.7111111111111111,"display":"71.1%","normalized":71.1,"normalizedMeaning":"71.1% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2Fmqsc4vFuKtAHXxbYyX9WwK.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath","value":0.2596491228070175,"display":"26.0%","normalized":27.7,"normalizedMeaning":"27.7% of the best (93.7%)","bestValue":0.936842105263158,"fieldSize":61,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F8o2J9uQLKDCTiCFwYdkJRK.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"frontiermath-tier-4","value":0,"display":"0.0%","normalized":0,"normalizedMeaning":"0.0% of the best (97.6%)","bestValue":0.976,"fieldSize":51,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FjuNVfKBcYTwaKnk9GRA2zL.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"proofbench","value":0.13,"display":"13.0%","normalized":13,"normalizedMeaning":"13.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"reasoning effort: high"},{"benchmark":"arena-vision","value":1266.419775373325,"display":"1,266","normalized":86.7,"normalizedMeaning":"43% expected win rate against the board leader (1,313)","bestValue":1313.059170076326,"fieldSize":54,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/vision","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-08-27"},{"benchmark":"arena-text","value":1456.9255681285117,"display":"1,457","normalized":85.6,"normalizedMeaning":"43% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"nemotron-3-super","name":"Nemotron 3 Super","lab":"NVIDIA","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":29.6,"indexRange":{"min":25.3,"max":49.5},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":true,"firstParty":false,"releaseDate":"2026-03-11","url":"https://opencharts.com/benchmarks/nemotron-3-super","categories":[{"category":"reasoning","mean":9.7,"benchmarkCount":1,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":49.5,"benchmarkCount":2,"coveredCount":7,"indexed":true,"counted":false}],"scores":[{"benchmark":"critpt","value":0.0314285714285714,"display":"3.1%","normalized":9.7,"normalizedMeaning":"9.7% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"scicode","value":0.35995370370370405,"display":"36.0%","normalized":58,"normalizedMeaning":"58.0% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"weirdml","value":0.3801,"display":"38.0%","normalized":40.9,"normalizedMeaning":"40.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"minimax-m2-5","name":"MiniMax M2.5","lab":"MiniMax","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":28.9,"indexRange":{"min":19.3,"max":35.2},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":27.8,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2026-02-12","url":"https://opencharts.com/benchmarks/minimax-m2-5","categories":[{"category":"reasoning","mean":34.8,"benchmarkCount":2,"coveredCount":6,"indexed":true,"counted":false},{"category":"coding","mean":17,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":4,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":21.2,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":67.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arc-agi-2","value":0.0486,"display":"4.9%","normalized":5.1,"normalizedMeaning":"5.1% of the best (95.0%)","bestValue":0.95,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arc-agi","value":0.6367,"display":"63.7%","normalized":64.6,"normalizedMeaning":"64.6% of the best (98.5%)","bestValue":0.985,"fieldSize":51,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://arcprize.org/leaderboard","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-webdev","value":1383.9560061233678,"display":"1,384","normalized":17,"normalizedMeaning":"9% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.04,"display":"4.0%","normalized":4,"normalizedMeaning":"4.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"terminal-bench","value":0.4270000000000001,"display":"42.7%","normalized":50.4,"normalizedMeaning":"50.4% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: cchuter"},{"benchmark":"apex-agents","value":0.062000000000000006,"display":"6.2%","normalized":13.1,"normalizedMeaning":"13.1% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":-23.15999999999999,"display":"$-23","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1390.544675228802,"display":"1,391","normalized":67.6,"normalizedMeaning":"34% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-5-1-codex","name":"GPT-5.1 Codex","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":28.8,"indexRange":{"min":26.5,"max":38.7},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":74.6,"hardSetCount":2,"openWeights":false,"firstParty":false,"releaseDate":"2025-11-13","url":"https://opencharts.com/benchmarks/gpt-5-1-codex","categories":[{"category":"coding","mean":13.1,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":9,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":64.3,"benchmarkCount":3,"coveredCount":8,"indexed":true,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1335.8372334802202,"display":"1,336","normalized":13.1,"normalizedMeaning":"7% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"proofbench","value":0.09,"display":"9.0%","normalized":9,"normalizedMeaning":"9.0% of the best (100.0%)","bestValue":1,"fieldSize":61,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.vals.ai/benchmarks/proof_bench","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · reasoning effort: high"},{"benchmark":"terminal-bench","value":0.604,"display":"60.4%","normalized":71.3,"normalizedMeaning":"71.3% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max · agent: Codex CLI"},{"benchmark":"metr-time-horizons","value":223.714694,"display":"3.7 h","normalized":77.8,"normalizedMeaning":"77.8% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"max"},{"benchmark":"apex-agents","value":0.207,"display":"20.7%","normalized":43.7,"normalizedMeaning":"43.7% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}}]},{"slug":"minimax-m2","name":"MiniMax M2","lab":"MiniMax","country":{"code":"CN","name":"China"},"rank":null,"rankRange":null,"index":28.3,"indexRange":{"min":14.2,"max":37.2},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":35.4,"hardSetCount":1,"openWeights":true,"firstParty":false,"releaseDate":"2025-10-27","url":"https://opencharts.com/benchmarks/minimax-m2","categories":[{"category":"coding","mean":10.7,"benchmarkCount":1,"coveredCount":7,"indexed":true,"counted":false},{"category":"agentic","mean":17.7,"benchmarkCount":2,"coveredCount":8,"indexed":true,"counted":false},{"category":"human-preference","mean":56.6,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"arena-webdev","value":1297.54091104803,"display":"1,298","normalized":10.7,"normalizedMeaning":"5% expected win rate against the board leader (1,797)","bestValue":1797.0426317593628,"fieldSize":86,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/webdev","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-05"},{"benchmark":"terminal-bench","value":0.3,"display":"30.0%","normalized":35.4,"normalizedMeaning":"35.4% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"vending-bench-2","value":160.5959999999999,"display":"$161","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1345.6777235258637,"display":"1,346","normalized":56.6,"normalizedMeaning":"28% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"gpt-oss-120b","name":"GPT-OSS 120B","lab":"OpenAI","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":27.8,"indexRange":{"min":21.5,"max":34.1},"grade":"F","eligible":false,"verification":"independent","hardSetScore":null,"hardSetMean":38,"hardSetCount":2,"openWeights":true,"firstParty":false,"releaseDate":"2025-08-05","url":"https://opencharts.com/benchmarks/gpt-oss-120b","categories":[{"category":"reasoning","mean":34.1,"benchmarkCount":3,"coveredCount":6,"indexed":true,"counted":true},{"category":"coding","mean":54,"benchmarkCount":3,"coveredCount":7,"indexed":true,"counted":false},{"category":"math","mean":88.9,"benchmarkCount":1,"coveredCount":5,"indexed":true,"counted":false},{"category":"agentic","mean":21.5,"benchmarkCount":4,"coveredCount":8,"indexed":true,"counted":true},{"category":"human-preference","mean":58.2,"benchmarkCount":1,"coveredCount":1,"indexed":false,"counted":false}],"scores":[{"benchmark":"gpqa-diamond","value":0.7575757575757576,"display":"75.8%","normalized":71.7,"normalizedMeaning":"71.7% of the best (95.8%), above chance","bestValue":0.9577020202020202,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2FLC5Rhog6wt2PT84CtJ483h.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"simplebench","value":0.221,"display":"22.1%","normalized":27,"normalizedMeaning":"27.0% of the best (81.9%)","bestValue":0.8190000000000001,"fieldSize":46,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://simple-bench.com/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"critpt","value":0.0114285714285714,"display":"1.1%","normalized":3.5,"normalizedMeaning":"3.5% of the best (32.3%)","bestValue":0.323,"fieldSize":78,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/critpt","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"aider-polyglot","value":41.8,"display":"41.8%","normalized":47.5,"normalizedMeaning":"47.5% of the best (88.0%)","bestValue":88,"fieldSize":5,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://aider.chat/docs/leaderboards/","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"scicode","value":0.388888888888889,"display":"38.9%","normalized":62.7,"normalizedMeaning":"62.7% of the best (62.0%)","bestValue":0.62037037037037,"fieldSize":73,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://artificialanalysis.ai/evaluations/scicode","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"weirdml","value":0.4817,"display":"48.2%","normalized":51.9,"normalizedMeaning":"51.9% of the best (92.9%)","bestValue":0.929,"fieldSize":69,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://htihle.github.io/weirdml.html","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"high"},{"benchmark":"aime-2024-2025","value":0.8888888888888888,"display":"88.9%","normalized":88.9,"normalizedMeaning":"88.9% of the best (100.0%)","bestValue":1,"fieldSize":78,"counted":true,"verification":"independent","provenance":"epoch-run","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://logs.epoch.ai/inspect-viewer/36231d6d/viewer.html?log_file=https%3A%2F%2Flogs.epoch.ai%2Finspect_ai_logs%2F3QjP5HPi8tKUzHtNJktfv4.eval","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"epoch-run"},"variantNote":"high"},{"benchmark":"terminal-bench","value":0.187,"display":"18.7%","normalized":22.1,"normalizedMeaning":"22.1% of the best (84.7%)","bestValue":0.8470000000000001,"fieldSize":40,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://www.tbench.ai/leaderboard/terminal-bench/2.0","license":"Apache 2.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"agent: Terminus 2"},{"benchmark":"metr-time-horizons","value":41.983451,"display":"42 min","normalized":53.8,"normalizedMeaning":"53.8% of the best (17.4 h) on a log scale","bestValue":1044.780145,"fieldSize":14,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"apex-agents","value":0.047,"display":"4.7%","normalized":9.9,"normalizedMeaning":"9.9% of the best (47.4%)","bestValue":0.474,"fieldSize":49,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://epoch.ai/benchmarks/apex-agents","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"vending-bench-2","value":-21.532000000000053,"display":"$-22","normalized":0,"normalizedMeaning":"0.0% of the best ($11,182) on a log scale","bestValue":11181.87333333333,"fieldSize":53,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Epoch AI — AI Benchmarking Hub","url":"https://andonlabs.com/evals/vending-bench-2","license":"CC BY 4.0 (via Epoch AI)","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"}},{"benchmark":"arena-text","value":1352.447968129115,"display":"1,352","normalized":58.2,"normalizedMeaning":"29% expected win rate against the board leader (1,507)","bestValue":1507.164171675996,"fieldSize":92,"counted":true,"verification":"independent","provenance":"leaderboard","source":{"publisher":"Arena (LMArena) — Leaderboard Dataset","url":"https://arena.ai/leaderboard/text","license":"CC BY 4.0","capturedAt":"2026-09-08T17:08:58.223Z","provenance":"leaderboard"},"variantNote":"board 2026-09-02"}]},{"slug":"theoarca-magnus-5-1","name":"TheoArca Magnus 5.1","lab":"TheoVex","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":null,"indexRange":null,"grade":null,"eligible":false,"verification":"none","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":true,"releaseDate":"2026-07-25","url":"https://opencharts.com/benchmarks/theoarca-magnus-5-1","categories":[],"scores":[]},{"slug":"theoarca-velox-5-1","name":"TheoArca Velox 5.1","lab":"TheoVex","country":{"code":"US","name":"United States"},"rank":null,"rankRange":null,"index":null,"indexRange":null,"grade":null,"eligible":false,"verification":"none","hardSetScore":null,"hardSetMean":null,"hardSetCount":0,"openWeights":false,"firstParty":true,"releaseDate":"2026-07-25","url":"https://opencharts.com/benchmarks/theoarca-velox-5-1","categories":[],"scores":[]}]}