MiniMax M2.1: benchmarks, price and rank
MiniMax M2.1 from MiniMax is 99th of 123 rated AI models on Spezia's leaderboard, strongest on MMLU-Pro. It costs $0.30 in, $1.20 out per million tokens.
- 99th of 123 rated models
- Rating 788
- Rated on 9 benchmarks
- Open weights
- Released Dec 23, 2025
- 1 million tokens of context
Where it is strongest
- 4th of 56 on MMLU-Pro
Where it falls behind
- 66th of 68 on SimpleBench
- 90th of 107 on LMArena text
- 76th of 98 on LMArena WebDev
Price: $0.30 in, $1.20 out per million tokens.
Its headline results
The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.
- BrowseComp62%
Questions whose answers are hidden deep on the web and take many searches and pages to track down.
36th of 56. The maker's own number, from MiniMax, via LLM Stats.
- Humanity's Last Exam22%
About 2,500 questions written by experts in more than a hundred fields to be as hard as they could make them, many needing a diagram or image.
50th of 74. The maker's own number, from MiniMax, via LLM Stats.
- Toolathlon43.5%
Long tasks across many real apps and their tools, such as email, calendars, spreadsheets, code repositories and online stores.
26th of 37. The maker's own number, from MiniMax, via LLM Stats.
- SimpleBench34.7%
Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.
66th of 68. Independent result from Epoch AI.
- AA-LCR62%
Questions that need reasoning across several long documents at once, about 100,000 tokens of text in all.
13th of 19. The maker's own number, from MiniMax, via LLM Stats.
- LMArena text1,392
People put a question to two anonymous models and vote for the better answer.
90th of 107. Independent result from LMArena.
- LMArena WebDev1,384
People ask two anonymous models to build the same web page or app and vote for the better one.
76th of 98. Independent result from LMArena.
- SciCode39%
Research coding problems written by scientists across 16 fields, from physics and chemistry to biology and materials science.
17th of 24. The maker's own number, from MiniMax, via LLM Stats.
- SWE-bench Verified67%
Real bug reports and feature requests from popular open-source Python projects. The model has to change the code so the project's hidden tests pass.
42nd of 56. The maker's own number, from MiniMax, via LLM Stats.
- GPQA Diamond81%
Graduate-level questions in biology, physics and chemistry, written by PhDs so that searching the web does not give the answer away.
The maker's own number, from MiniMax, via LLM Stats.
3 more results in its full profile.
Every result with its source, what it costs per task, what jumped with its release, side by side with any other model, and answers to your questions about when to use it.
Models ranked near it
How the rating works, and every benchmark it uses: the model leaderboard.