OpenAIUnited States· released Aug 2025· open weights
Provisional: 12 comparable tests so far. Ranked once it reaches 6 across 3 counting categories.
OpenAI, United States. One of 26 models from this lab on the Index.
Available in Theo as Theo Open Core
Theo orchestrates it for the steps it does best and always shows which engine answered.
Snapshot September 8, 2026
OpenCharts Index
27.8
Provisional: not enough comparable tests or counting categories to rank yet · 12 comparable tests · 2 counting categories
Drop any single test and the Index lands between 21.5 and 34.1.
Needs more results (2 of 8 so far)
Averaging 38.0 on the 2 it has; the score opens at 4 of 8.
Distance to today's frontier on the hardest tests. Not a measure of intelligence.
What it's known for
Strongest in reasoning (34.1) and agentic (21.5). It trails the frontier most in agentic, 78.5% behind on average. No comparable results yet in knowledge, multimodal and long context. Below 40% of the best published results, on average.
Strongest category
Reasoning
Hard, novel problems: graduate science, abstract puzzles, expert exams.
34.1
Mean of 3 of the category's 6 comparable tests. Counts toward the Index.
Best single result
Solve competition mathematics problems that have a single exact integer answer.
88.9%
88.9% of the best (100.0%) · #41 of 78.
Olympiad-style problems need long, careful chains of reasoning, and there is no partial credit.
Where it trails
Run a simulated vending business for a year: ordering, pricing and cash flow. The score is the final balance.
$-22
100.0% behind the best published result ($11,182) on a log scale · #44 of 53.
Long-horizon decisions whose consequences compound. It exposes models that drift or forget.
Not measured yet
No comparable result yet in these 3 categories. A blank is a blank, never a zero, and it does not lower the Index.
Covered, but not averaged into the Index:
See it at work
Every test it has taken, by category, with the kind of task it faced, who produced the number and where the result landed against the best published one. The examples are original and illustrative, never items from the datasets themselves.
Math · mean 88.9 across 1 comparable test · not counted toward the Index yet (fewer than half the category's tests)
Solve competition mathematics problems that have a single exact integer answer.
88.9%
88.9% of the best (100.0%)
#41 of 78 · n = 45
high
45 problems: one problem is 2.2 points, so gaps of a few points are within noise.
Source: Epoch AI (internal runs)Every result is read against the best published one on its test: percent scores as a share of the best above chance, Arena ratings as the expected win rate against the board leader, open-ended values on a log scale. Sitting a harder exam never lowers a model. Every number links to the publisher that produced it.
Head to head
Choose a rival. Every counted test both have taken appears side by side as a share of the best published result on that test, with the category means above. A win is a higher share; a gap under 2 points is a tie, because one item on a 45-problem exam is 2.2 points and publishers' own standard errors are of that order.
GPT-OSS 120B
provisional·Index 27.8F
0 : 6
wins · 0 ties · 6 shared
GPT-6 Astra
#1·Index 99.0A+
GPT-6 Astra comes out ahead on 6 of the 6 tests both have taken.
Category means
Reasoning
Knowledge
Coding
Math
Agentic
Human preference
Shared tests · biggest gaps first
Bars are distances to the best published result on each test (percent scores above chance, Arena ratings as win rate against the leader, open-ended values on a log scale). Raw values as the publishers report them.
Open GPT-6 AstraScores
The best published run per benchmark, the raw value as the publisher reports it, who produced the number, how close it comes to the best published result on that test, and where the model sits among every model scored on it. Percent scores are read above chance, Arena ratings as a win rate against the board leader, open-ended values on a log scale. A test with too few models to compare against is shown but not counted.
Category means
Mean · tests taken of the category's comparable tests. A faded category either rests on fewer than half of them or is a single-test category (marked “beside”), which is shown next to the Index and never averaged into it.
75.8%
71.7% of the best (95.8%), above chance
8th percentile of 78 · n = 198
high
Source: Epoch AI — AI Benchmarking Hub22.1%
27.0% of the best (81.9%)
0th percentile of 46
1.1%
3.5% of the best (32.3%)
18th percentile of 78
high
Source: Epoch AI — AI Benchmarking Hub41.8%
47.5% of the best (88.0%)
0th percentile of 5 · n = 225
high
Source: Epoch AI — AI Benchmarking Hub38.9%
62.7% of the best (62.0%)
13th percentile of 73 · n = 338
high
Source: Epoch AI — AI Benchmarking Hub48.2%
51.9% of the best (92.9%)
31th percentile of 69
high
Source: Epoch AI — AI Benchmarking Hub88.9%
88.9% of the best (100.0%)
43th percentile of 78 · n = 45
high
Source: Epoch AI — AI Benchmarking Hub18.7%
22.1% of the best (84.7%)
8th percentile of 40
agent: Terminus 2
Source: Epoch AI — AI Benchmarking Hub42 min
53.8% of the best (17.4 h) on a log scale
0th percentile of 14
4.7%
9.9% of the best (47.4%)
6th percentile of 49
$-22
0.0% of the best ($11,182) on a log scale
0th percentile of 53
1,352
29% expected win rate against the board leader (1,507)
3th percentile of 92
board 2026-09-02
Source: Arena (LMArena) — Leaderboard DatasetTop of the leaderboard
28 of the models on this page run inside Theo today. Theo picks the right one for each step and always shows which engine answered.