Back to Models
Zhipu AIOpen Source
GLM-4.7
GLM-4.7 is a large-scale foundation model developed by Zhipu AI. It boasts an extremely high scale of approximately 3.58 trillion parameters and is a chat-focused model with a long context window of 200K.
Parameters
3580.0B
Context Window
200K
License
MIT
Release Date
2025-12-22
API Pricing
API pricing for this model is not yet available
Strengths
- ・Massive 3.58 trillion parameters
- ・Supports long 200K token context
- ・Openness via MIT license
Weaknesses
- ・Extremely large model size
- ・Requires high computational resources
- ・Potentially high operational costs
Use Cases
- ・Building advanced conversational AI services
- ・Analyzing and processing long documents
- ・Research and development of large language models
Deep Analysis
Parameters
3.58T MoE
Context Window
200K tokens
Tau2-Bench
84.7
Open-source SOTA
BrowseComp
67
Pricing
~$0.50/$1.60 per 1M tokens
Release Date
Late 2025
Strengths
- ・Strong agentic coding with think-then-act
- ・Tau2-Bench 84.7 open-source SOTA
- ・Improved frontend aesthetics
- ・200K context with 128K output
Weaknesses
- ・Text-only
- ・Superseded by GLM-5 series
- ・3.58T params extremely expensive to self-host
Competitor Comparison
| Model | Price | ||
|---|---|---|---|
| Claude Sonnet 4.5 | $3/$15 | ||
| DeepSeek V3 | $0.27/$1.10 | ||
| GLM-5 | 1452 | 77.8 | $1/$3.20 |
Sources
Analysis generated: 2026-05-24