Loading page content
GLM-5.2Open
Tool use · Reasoning
Evaluation overview
Overall rank
#11
88 points · 180 models
Open-weight rank
#1
100 points · 109 models
Data coverage
66 items
7 sources · updated through 2026-07-17
Domain performance
Coding#9 / 160
91.1
Research#13 / 174
93.6
Agents#11 / 125
88.2
Long context#13 / 82
85.8
Evaluation highlights
All evaluationsBasic information
- Context (input / output)
- 1M / 131K
- Parameters
- 744B / 40B active
- Release date
- 2026-06
- License
- MIT
- Input modalities
- Text
- Output modalities
- Text
- Context
- 1M
- Architecture
- MoE (256 experts), DSA
- Active parameters
- 40B
Availability
Zhipu AIFull pricing
- Input
- ≈ $1.1429 /1M
- Output
- ≈ $4 /1M
vLLM · FP8Deployment options
FP8 · 8 × 141 GB VRAM
Related links
Version relationships
GLM-5 family
- Runtime modes
- Max