Kimi-K2.5
Moonshot AI- Arena score
- 1445
- Among open models
- #13of 148
- Among all on Arena
- #67
- Votes
- 73,820
Measured on Kimi K2.5 (Thinking), the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
Top 10 among open models in#9Coding#10Applied math
CategoryScoreOpen rankArena rankVotes
- Chinese1493#17#664.5K
- Coding1484#9#501.9K
- Applied math1473#10#504K
- Software & IT1471#11#622.9K
- Math1469#12#484.1K
- Expert1465#14#696.9K
- Hard prompts, English1460#13#6820.8K
- Science1458#15#6411.6K
- French1454#20#772.6K
- English1453#13#6231.4K
- Hard prompts1453#12#7047.3K
- Medicine & health1449#23#795.1K
- Spanish1449#17#662.5K
- Polish1445#13#611.5K
- Multi-turn1444#17#8113K
- Long prompts1444#15#7529.3K
- Excluding ties1442#12#6453.8K
- German1441#15#701.3K
- Legal & government1441#21#895.6K
- Russian1435#15#778.7K
- Non-English1432#12#6742.4K
- Instructions1431#14#7523.9K
- Business & finance1427#24#9014.2K
- Writing & literature1425#13#7517.7K
- Japanese1422#12#53834
- Creative writing1422#12#6812.1K
- Media & sports1417#12#6915.3K
- Korean1410#11#581.3K
About as strong, smaller
Compare
gemma-4-26B-A4B-itGoogle143425.8B

Qwen3.8-27BQwen143727.8B

gemma-4-31B-itGoogle144331.3B

DeepSeek-V4-FlashDeepSeek1432291B
Discussion
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiounsloth/Kimi-K2.5-GGUF12 builds
- MLX · Apple silicon: mlx-lm, LM Studiomlx-community/Kimi-K2.56 builds
Most downloaded
BenchClub runs
Not runnable yet
Runs open once a Kimi-K2.5 build is registered.
More from Moonshot AI
See more
Kimi-K3768.9K2.8T

Kimi-K2.6202.7K1.0T

Kimi-K3.1-Pro53.2K2.8T

Kimi-K2-Instruct252.1K1.0T
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