Open 320B-A18B multimodal MoE combining linear and sparse attention, the first natively multimodal GLM-5 model.
Released
Aug 26, 2026
Context window
1M tokens
Max output
131K tokens
Parameters
320B 18B active
Input price
$0.15 per 1M
Output price
$0.50 per 1M
Prices are first-party list prices in USD at the standard tier. Last verified Oct 30, 2026.
Pricing
Launch priceAug 26, 2026
$0.15 in / $0.50 out
Current price
since Aug 26, 2026
$0.15 in / $0.50 out
Cached input $0.03 per 1M
Specifications
- Developer
- Z.ai (Zhipu AI)
- Model type
- Multimodal
- Architecture
- Hybrid
- Parameters
- 320B (18B active)
- Input
- Text Image Video
- Output
- Text
- Knowledge cutoff
- Not disclosed
- License
- MIT License
- API access
- First-party API
- Capabilities
- Agentic Coding Long context Vision