Text ArenaLeaderboard
Answer the same prompt as a rival model; real users vote blind on which answer they prefer.
1,384
33% expected win rate against the board leader (1,507)
#88 of 92
preview · board 2026-09-02
Source: ArenaMiniMaxChina· released Dec 2025· open weights
MiniMax M2.1
Provisional: 3 comparable tests so far. Ranked once it reaches 6 across 3 counting categories.
MiniMax, China. One of 5 models from this lab on the Index.
Third-party publishedLow confidence (thin coverage)
Snapshot September 8, 2026
OpenCharts Index
42.1
Provisional: not enough comparable tests or counting categories to rank yet · 3 comparable tests · 0 counting categories
Drop any single test and the Index lands between 30.3 and 54.6.
Needs more results (1 of 8 so far)
Averaging 43.2 on the 1 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
MiniMax M2.1, in 3 results.
Its widest gap is on Code Arena (WebDev), 82.7% behind the best published result. No comparable results yet in reasoning, knowledge, math, multimodal and long context. Between 50% and 60% behind the best published results, on average.
Strongest category
No category counts yet: a category counts once it has more than one comparable test and the model has taken at least half of them. The scores it has are below.
Best single result
Text Arena
Answer the same prompt as a rival model; real users vote blind on which answer they prefer.
1,384
33% expected win rate against the board leader (1,507) · #88 of 92.
Millions of blind votes on real prompts are the market's measure of what people prefer.
Where it trails
Code Arena (WebDev)
Build a web app from the same prompt as a rival model; real users vote blind on the result.
1,387
82.7 points short of parity with the board leader (1,797) · #62 of 86.
People judging finished apps side by side is the most honest measure of front-end quality.
Not measured yet
No comparable result yet in these 5 categories. A blank is a blank, never a zero, and it does not lower the Index.
- Reasoning
- Knowledge
- Math
- Multimodal
- Long context
Covered, but not averaged into the Index:
- Coding1 of 7 tests · needs 4 · mean 17.3
- Agentic1 of 8 tests · needs 4 · mean 43.2
- Human preferencesingle-test category, shown beside the Index · 65.9
See it at work
What MiniMax M2.1 was asked to do, and how it did.
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.
Human preference · mean 65.9 across 1 comparable test · a single-test category, shown beside the Index and never averaged into it
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
MiniMax M2.1 against whoever you pick.
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.
MiniMax M2.1
provisional·Index 42.1D
0 : 3
wins · 0 ties · 3 shared
GPT-5.5
#7·Index 80.2B+
GPT-5.5 comes out ahead on 3 of the 3 tests both have taken.
Category means
Reasoning
—89.4Knowledge
—83.3Coding
17.374.8Math
—74.9Agentic
43.281.5Multimodal
—92.3Human preference
65.992.8
Shared tests · biggest gaps first
- Terminal-BenchAgentic−56.836.6%84.7%
- Text ArenaHuman preference−26.91,3841,482
- Code Arena (WebDev)Coding−14.81,3871,510
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-5.5Scores
Every test, every source
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
- Reasoning—
- Knowledge—
- Coding17.3 · 1/7
- Math—
- Agentic43.2 · 1/8
- Multimodal—
- Human preferencebeside65.9 · 1/1
- Long context—
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.
Code Arena (WebDev)LeaderboardCoding
1,387
9% expected win rate against the board leader (1,797)
28th percentile of 86
preview · board 2026-09-05
Source: Arena (LMArena) — Leaderboard DatasetTerminal-BenchLeaderboardAgentic
36.6%
43.2% of the best (84.7%)
36th percentile of 40
agent: Crux
Source: Epoch AI — AI Benchmarking HubText ArenaLeaderboardHuman preference
1,384
33% expected win rate against the board leader (1,507)
4th percentile of 92
preview · board 2026-09-02
Source: Arena (LMArena) — Leaderboard DatasetTop of the leaderboard
Compare with its neighbours.
The best models, ranked here, working inside Theo.
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.