Gemini 4 Argon
Claude Opus 5.5
Muse Wonder
MiMo-V2.6-Pro
Qwen3.8-Max
Kimi K3
GLM-5.3
GPT-5.5
ERNIE 5.1
DeepSeek-V4.1-Flash-20260910
Grok 4.20 Beta
amazon-nova-experimental-chat
Dola Seed 2.0 Pro
NVIDIA-Nemotron-3-Ultra-550B-A55B
Gemma 4 31B
Step 5 Preview
MiniMax-M2.1 Preview
MiniMax-M2.1 Preview is an open-weight model from MiniMax, released Dec 23, 2025. It ranks #167 of 351 on the LMArena text leaderboard with a rating of 1392. It costs $0.30 per million input tokens and $1.20 per million output tokens, with a 200K-token context window.
- Developer
- MiniMax
- Released
- Dec 23, 2025
- Context window
- 200K
- Price in / out, per 1M tokens
- $0.30 / $1.20
- Weights
- Open · MIT
- Inputs
- Text
- Reasoning
- Yes
- Tool use
- Yes
- Hugging Face downloads, 30 days
- 12.1K
- Hugging Face likes
- 1.4K
The rating LMArena published at each update since Dec 30, 2025. Ratings move as votes come in and new models join.
Benchmarks
| Benchmark | Score bar | Accuracy | Rank | Cost / test |
|---|---|---|---|---|
| SWE-benchCoding | 74.80% | 44 / 85 | $0.44 | |
| LiveCodeBenchCoding | 81.76% | 60 / 143 | $0.3 / $1.2 | |
| MedCodeMedicine | 34.08% | 82 / 107 | $0.3 / $1.2 | |
| MedQAMedicine | 91.16% | 41 / 92 | $0.3 / $1.2 | |
| GPQA DiamondReasoning | 78.54% | 75 / 138 | $0.3 / $1.2 | |
| MMLU-ProReasoning | 87.05% | 35 / 138 | $0.3 / $1.2 | |
| AIMEReasoning | 77.92% | 49 / 96 | $0.3 / $1.2 |
Vals AI, as of Oct 24, 2026.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1392 | 167 / 351 | 17.2K |
| CodingLMArena | 1421 | 149 / 346 | 3.5K |
| MathLMArena | 1396 | 168 / 339 | 1K |
| Creative writingLMArena | 1362 | 156 / 349 | 2.8K |
| Instruction followingLMArena | 1400 | 130 / 351 | 4.7K |
| Hard promptsLMArena | 1411 | 150 / 351 | 9.3K |
| Multi-turnLMArena | 1397 | 157 / 349 | 2.8K |
| Long queriesLMArena | 1416 | 118 / 330 | 4.5K |
| ExpertLMArena | 1431 | 119 / 305 | 1.2K |
| Legal and governmentIndustries | 1410 | 145 / 325 | 1.2K |
| Medicine and healthcareIndustries | 1387 | 175 / 325 | 1.2K |
| Business and financeIndustries | 1390 | 156 / 344 | 3.2K |
| ChineseLanguages | 1430 | 152 / 328 | 1.4K |
| SpanishLanguages | 1396 | 151 / 249 | 489 |
| GermanLanguages | 1381 | 142 / 261 | 297 |
| FrenchLanguages | 1404 | 151 / 252 | 446 |
| JapaneseLanguages | 1287 | 168 / 236 | 178 |
| WebDevLMArena | 1385 | 108 / 124 | 7.6K |
LMArena, CC BY 4.0, as of Oct 22, 2026. Where a model is rated at several effort levels, its best is shown.
Compare MiniMax-M2.1 Preview
MiniMax
Builds long-context text, voice and video models, with open-weight releases. Founded 2021, based in Shanghai. minimax.io
Questions
Who makes MiniMax-M2.1 Preview?
MiniMax-M2.1 Preview is made by MiniMax, founded 2021 and based in Shanghai.
How much does MiniMax-M2.1 Preview cost?
$0.30 per million input tokens and $1.20 per million output tokens, at its provider’s list price.
What is MiniMax-M2.1 Preview’s context window?
200K tokens (204,800).
How does MiniMax-M2.1 Preview rank?
#167 of 351 on LMArena’s text leaderboard, rated 1392 (95% interval 1387–1397) from 17,230 votes, as of Oct 22, 2026.
Is MiniMax-M2.1 Preview open source?
MiniMax-M2.1 Preview has open weights under the MIT license, downloadable from Hugging Face (MiniMaxAI/MiniMax-M2.1).