GLM-4.6
Z.ai- Arena score
- 1440
- Among open models
- #16of 148
- Among all on Arena
- #77
- Votes
- 35,631
Measured on GLM-4.6, the served version, not a local build. LMArena, CC BY 4.0, published Nov 11, 2026.
Score by category
CategoryScoreOpen rankArena rankVotes
- Chinese1500#11#552.7K
- French1461#14#62702
- Science1456#18#715.9K
- Software & IT1455#25#9312.6K
- Medicine & health1455#18#651.9K
- English1449#16#7615.2K
- Coding1449#32#1087.6K
- Legal & government1448#15#752.4K
- Applied math1448#23#821.8K
- Polish1447#11#581.3K
- German1447#13#65681
- Hard prompts, English1445#26#1008.2K
- Hard prompts1440#24#9719.3K
- Excluding ties1437#16#7725K
- Spanish1437#23#881K
- Expert1433#34#1152.1K
- Math1432#26#972.1K
- Multi-turn1427#26#1095.6K
- Non-English1426#18#8320.4K
- Long prompts1422#27#1069.1K
- Russian1418#25#1053.5K
- Business & finance1416#33#1136.9K
- Writing & literature1416#15#858K
- Creative writing1412#14#815.1K
- Instructions1410#27#10610.2K
- Media & sports1409#13#806.6K
- Korean1400#14#72612
- Japanese1394#22#90372
About as strong, smaller
Compare
gemma-4-26B-A4B-itGoogle143425.8B

Qwen3.8-27BQwen143727.8B

gemma-4-31B-itGoogle144331.3B
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiounsloth/GLM-4.6-GGUF16 builds
- MLX · Apple silicon: mlx-lm, LM Studiomlx-community/GLM-4.6-4bit7 builds
- AWQ · vLLM, SGLangbullpoint/GLM-4.6-AWQ1 build
- GPTQ · vLLM, SGLangModelCloud/GLM-4.6-GPTQMODEL-W4A16-v23 builds
Most downloaded
BenchClub runs
Not runnable yet
Runs open once a GLM-4.6 build is registered.
More from Z.ai
See more
GLM-5.3-Flash6.4M321B
GLM-5.31.7M753B
GLM-51.2M754B
GLM-4.7-Flash1.2M31.2B
Other >300B models
See more
DeepSeek-V3.2DeepSeek9.6M685B
InklingThinking Machines1.5M952B
NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1M561B

Kimi-K3Moonshot AI768.9K2.8T