MiMo-V2.5
Xiaomi MiMo- Arena score
- 1427
- Among open models
- #27of 148
- Among all on Arena
- #102
- Votes
- 46,793
Measured on MiMo-V2.5, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Chinese1470#24#882.5K
- Coding1469#13#7113.5K
- Software & IT1463#17#7519.1K
- Hard prompts, English1461#12#6614.9K
- Expert1460#21#774.7K
- French1450#22#821.8K
- Hard prompts1450#16#7731.4K
- Applied math1449#22#802.6K
- English1447#18#8221.1K
- Multi-turn1445#15#798.3K
- Long prompts1445#14#7321K
- Science1439#30#1067.6K
- Math1436#24#892.4K
- Instructions1434#11#7015.9K
- Business & finance1432#17#769.6K
- Medicine & health1431#34#1133.4K
- Legal & government1423#35#1273.7K
- German1419#27#102802
- Excluding ties1417#27#10335.9K
- Spanish1416#39#1271.4K
- Polish1410#31#114935
- Writing & literature1408#22#10111.4K
- Non-English1404#35#12625.7K
- Media & sports1399#24#10110.3K
- Russian1395#42#1465.2K
- Creative writing1391#29#1167.8K
- Korean1356#37#135773
- Japanese1305#45#162533
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
- Xiaomi MiMo-v2.5 Series API Permanent Price Reduction Up to 99%gainsurier · May 26, 2026134 points
- Inference Optimization for MiMo v2.5: Pushing Hybrid SWA Efficiency to the Limittheanonymousone · Jul 7, 2026116 points
- Xiaomi releases MiMo-v2.5 Family weights with strong coding and agent benchmarkssteveharing1 · Apr 28, 202621 points
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiohuihui-ai/Huihui-MiMo-V2.5-abliterated-GGUF17 builds
- MLX · Apple silicon: mlx-lm, LM Studiodawncr0w/MiMo-V2.5-oQ4-MLX3 builds
- AWQ · vLLM, SGLangspectator2026/MiMo-V2.5-AWQ-int43 builds
Most downloaded
BenchClub runs
Not runnable yet
Runs open once a MiMo-V2.5 build is registered.
More from Xiaomi MiMo
See more
MiMo-V2.6-Pro-RL24.9K1.0T
MiMo-V2-Flash16.1K310B
MiMo-V2.5-Pro9.1K1.0T
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