GLM-5.2
Z.ai- Arena score
- 1469
- Among open models
- #4of 148
- Among all on Arena
- #33
- Votes
- 54,143
Measured on GLM-5.2 (Max), 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#1Creative writing#2Spanish#3Legal & government#3Science
CategoryScoreOpen rankArena rankVotes
- Chinese1522#4#333.9K
- Legal & government1489#3#224.3K
- Applied math1489#5#352.9K
- Expert1488#8#445.5K
- Science1488#3#279K
- Coding1488#7#4114.4K
- Software & IT1487#6#3820.9K
- Math1483#4#282.4K
- Spanish1483#2#171.7K
- Hard prompts, English1481#8#3814.5K
- Hard prompts1480#6#4035.1K
- Long prompts1479#5#3324.5K
- French1476#5#372K
- Medicine & health1475#5#334K
- English1475#6#3522K
- Excluding ties1474#4#3440.2K
- Multi-turn1469#6#418.6K
- German1467#4#35991
- Instructions1467#7#3818.6K
- Russian1466#5#406.5K
- Writing & literature1461#3#3314.5K
- Polish1460#4#39904
- Creative writing1459#1#2610.9K
- Non-English1458#4#3632.2K
- Japanese1451#4#34892
- Business & finance1450#8#4710.3K
- Media & sports1448#3#3213.4K
- Korean1428#4#37978
About as strong, smaller
Compare
GLM-5.3-FlashZ.ai1472321B
Discussion
- Calibri runs 744B parameter AI on laptop without a GPU using GLM 5.2 #CalibriAI #GLM5 #AIDisruptionJordan Blake · Sep 12, 202659.4K likes
- GLM 5.2 beats Claude in our benchmarksjms703 · Jun 28, 20261.1K points
- Show HN: Getting GLM 5.2 running on my slow computervforno · Jul 9, 2026937 points
- GLM-5.2 is the new leading open weights model on Artificial Analysishimata4113 · Jun 17, 2026916 points
- GLM 5.2 Is Outaloknnikhil · Jun 13, 2026772 points
Built on it
Ways to run it
- GGUF · llama.cpp, Ollama, LM Studiounsloth/GLM-5.2-GGUF43 builds
- MLX · Apple silicon: mlx-lm, LM Studiomlx-community/GLM-5.2-DQ4plus-q822 builds
- GPTQ · vLLM, SGLangcanada-quant/GLM-5.2-W4A16-MTP2 builds
Most downloaded
BenchClub runs
Not runnable yet
Runs open once a GLM-5.2 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