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-M2.7
MiniMax-M2.7 is an open-weight model from MiniMax, released Mar 18, 2026. It ranks #149 of 351 on the LMArena text leaderboard with a rating of 1405. It costs $0.30 per million input tokens and $1.20 per million output tokens, with a 200K-token context window.
- Developer
- MiniMax
- Released
- Mar 18, 2026
- Context window
- 200K
- Price in / out, per 1M tokens
- $0.30 / $1.20
- Weights
- Open · Modified MIT
- Inputs
- Text
- Reasoning
- Yes
- Tool use
- Yes
- Hugging Face downloads, 30 days
- 438.1K
- Hugging Face likes
- 1.3K
The rating LMArena published at each update since Mar 18, 2026. Ratings move as votes come in and new models join.
Benchmarks
| Benchmark | Score bar | Accuracy | Rank | Cost / test |
|---|---|---|---|---|
| SWE-benchCoding | 73.80% | 48 / 85 | $0.47 | |
| Vibe Code BenchCoding | 11.93% | 96 / 115 | $2.82 | |
| LiveCodeBenchCoding | 79.93% | 72 / 143 | $0.3 / $1.2 | |
| Tax Agent BenchFinance | 6.87% | 69 / 76 | $0.12 | |
| Legal Research BenchLaw | 10.58% | 76 / 85 | $0.22 | |
| MedCodeMedicine | 34.44% | 80 / 107 | $0.3 / $1.2 | |
| GPQA DiamondReasoning | 86.62% | 42 / 138 | $0.3 / $1.2 | |
| MMLU-ProReasoning | 80.43% | 90 / 138 | $0.3 / $1.2 | |
| AIMEReasoning | 91.04% | 26 / 96 | $0.3 / $1.2 |
Vals AI, as of Oct 24, 2026.
LMArena, by category
| Leaderboard | Rating | Rank | Votes |
|---|---|---|---|
| TextLMArena | 1405 | 149 / 351 | 87.7K |
| CodingLMArena | 1454 | 93 / 346 | 24.8K |
| MathLMArena | 1421 | 122 / 339 | 4.4K |
| Creative writingLMArena | 1354 | 163 / 349 | 15.1K |
| Instruction followingLMArena | 1405 | 116 / 351 | 29.4K |
| Hard promptsLMArena | 1423 | 128 / 351 | 57K |
| Multi-turnLMArena | 1412 | 133 / 349 | 15K |
| Long queriesLMArena | 1419 | 113 / 330 | 37.3K |
| ExpertLMArena | 1443 | 93 / 305 | 8.8K |
| Legal and governmentIndustries | 1403 | 153 / 325 | 6.7K |
| Medicine and healthcareIndustries | 1409 | 148 / 325 | 6K |
| Business and financeIndustries | 1407 | 128 / 344 | 16.8K |
| ChineseLanguages | 1441 | 141 / 328 | 5.8K |
| SpanishLanguages | 1403 | 143 / 249 | 2.8K |
| GermanLanguages | 1398 | 125 / 261 | 1.5K |
| FrenchLanguages | 1424 | 129 / 252 | 3.3K |
| JapaneseLanguages | 1259 | 183 / 236 | 1.2K |
| WebDevLMArena | 1398 | 99 / 124 | 16K |
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-M2.7
MiniMax
Builds long-context text, voice and video models, with open-weight releases. Founded 2021, based in Shanghai. minimax.io
Questions
Who makes MiniMax-M2.7?
MiniMax-M2.7 is made by MiniMax, founded 2021 and based in Shanghai.
How much does MiniMax-M2.7 cost?
$0.30 per million input tokens and $1.20 per million output tokens, at its provider’s list price.
What is MiniMax-M2.7’s context window?
200K tokens (204,800).
How does MiniMax-M2.7 rank?
#149 of 351 on LMArena’s text leaderboard, rated 1405 (95% interval 1402–1408) from 87,731 votes, as of Oct 22, 2026.
Is MiniMax-M2.7 open source?
MiniMax-M2.7 has open weights under the Modified MIT license, downloadable from Hugging Face (MiniMaxAI/MiniMax-M2.7).