Moonshot AIChina· released Jun 2026· open weights
Ranked #29, 41.4% behind the frontier on average.
Moonshot AI, China. One of 5 models from this lab on the Index.
Snapshot September 8, 2026
OpenCharts Index
58.6
Ranked #29 (29 to 30) of 36 · 41.4% behind the frontier on average · 14 comparable tests · 3 counting categories
Drop any single test and the Index lands between 56.2 and 61.2, anywhere from #29 to #30. Neighbours inside that range are ties.
Needs more results (2 of 8 so far)
Averaging 35.1 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 (63.5) and coding (57.1). It trails the frontier most in math, 44.7% behind on average. No comparable results yet in multimodal, human preference and long context. Between 40% and 50% behind the best published results, on average. Drop any single test and it would sit anywhere from #29 to #30.
Strongest category
Reasoning
Hard, novel problems: graduate science, abstract puzzles, expert exams.
63.5
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.
95.6%
95.6% of the best (100.0%) · #19 of 78.
Olympiad-style problems need long, careful chains of reasoning, and there is no partial credit.
Where it trails
Solve the hardest FrontierMath tier: problems that take expert mathematicians days.
12.2%
87.5% behind the best published result (97.6%) · #40 of 51.
The deepest end of the set. Progress here signals genuinely new capability.
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.
Agentic · mean 66.4 across 2 comparable tests · not counted toward the Index yet (fewer than half the category's tests)
Run a simulated vending business for a year: ordering, pricing and cash flow. The score is the final balance.
$5,083
74.6% of the best ($11,182) on a log scale
#24 of 53
Andon Labs estimates a strong human operator at roughly $63,000, so every model is far from the ceiling.
Source: Andon LabsComplete professional-services work in consulting, law and finance as an agent, graded against expert rubrics.
27.6%
58.2% of the best (47.4%)
#22 of 49
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.
Kimi K2.7 Code
#29·Index 58.6C-
0 : 11
wins · 0 ties · 11 shared
GPT-6 Astra
#1·Index 99.0A+
GPT-6 Astra comes out ahead on 11 of the 11 tests both have taken.
Category means
Reasoning
Knowledge
Coding
Math
Agentic
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.
87.9%
88.8% of the best (95.8%), above chance
48th percentile of 78 · n = 198
57.9%
70.7% of the best (81.9%)
40th percentile of 46
10.0%
31.0% of the best (32.3%)
55th percentile of 78
36.5%
48.3% of the best (75.6%)
38th percentile of 57 · n = 1,000
47.5%
76.5% of the best (62.0%)
51th percentile of 73 · n = 338
54.1%
58.3% of the best (92.9%)
44th percentile of 69
30.1%
56.2% of the best (53.5%)
38th percentile of 27
harness: mini-swe-agent · reasoning effort: none
Source: Epoch AI — AI Benchmarking Hub49.7%
67.7% of the best (73.4%)
15th percentile of 21
1,472
13% expected win rate against the board leader (1,797)
54th percentile of 86
board 2026-09-05
Source: Arena (LMArena) — Leaderboard Dataset95.6%
95.6% of the best (100.0%)
71th percentile of 78 · n = 45
54.0%
57.7% of the best (93.7%)
43th percentile of 61 · n = 290
12.2%
12.5% of the best (97.6%)
18th percentile of 51 · n = 48
27.6%
58.2% of the best (47.4%)
54th percentile of 49
$5,083
74.6% of the best ($11,182) on a log scale
56th percentile of 53
Nearby on 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.