MiniMax-M3
MiniMax- Arena score
- 1428
- Among open models
- #25of 148
- Among all on Arena
- #100
- Votes
- 73,579
Measured on MiniMax-M3, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Coding1468#15#7320.5K
- Chinese1465#27#984.8K
- Software & IT1463#16#7429.2K
- Expert1460#20#767.8K
- Hard prompts, English1452#20#8320.2K
- Science1446#23#8911.7K
- Applied math1443#24#864.1K
- Hard prompts1442#23#9448.5K
- English1440#25#9829.7K
- Long prompts1440#18#8032.9K
- Multi-turn1439#20#9112.7K
- French1439#29#1052.7K
- Medicine & health1433#32#1084.9K
- Business & finance1431#19#8014K
- Legal & government1430#30#1135.9K
- German1429#22#931.3K
- Instructions1428#16#7925.5K
- Spanish1424#34#1102.3K
- Math1423#35#1153.7K
- Russian1422#23#999.1K
- Excluding ties1419#26#10155.7K
- Polish1417#29#1071.3K
- Non-English1414#25#10143.9K
- Writing & literature1411#18#9319.1K
- Media & sports1398#27#10517.4K
- Creative writing1395#27#11113.8K
- Korean1365#29#1181.2K
- Japanese1363#31#1191.1K
About as strong, smaller
Compare
gemma-4-26B-A4B-itGoogle143425.8B

Qwen3.8-27BQwen143727.8B

gemma-4-31B-itGoogle144331.3B

Qwen3-Next-80B-A3B-InstructQwen141781.3B
Discussion
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiounsloth/MiniMax-M3-GGUF16 builds
- MLX · Apple silicon: mlx-lm, LM Studiomlx-community/MiniMax-M3-4bit11 builds
- AWQ · vLLM, SGLangspectator2026/MiniMax-M3-AWQ-int41 build
- GPTQ · vLLM, SGLangSebesky/MiniMax-M3-W4A16-GPTQ2 builds
Most downloaded
BenchClub runs
Not runnable yet
Runs open once a MiniMax-M3 build is registered.
More from MiniMax
See more
MiniMax-M2.51M229B

MiniMax-M2.7222.2K229B

MiniMax-M2.113.9K229B

MiniMax-M2112.4K229B
Other >300B models
See more
DeepSeek-V3.2DeepSeek9.6M685B
GLM-5.3-FlashZ.ai6.4M321B
InklingThinking Machines1.5M952B
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1M561B