GLM-4.6
by Z.AI
· GLM family · released Sep 30, 2025
Open weights
Active
Zhipu/Z.ai's open-weight MoE flagship (357B total / 32B active) focused on coding, agentic use, and token efficiency. MIT-licensed, 200K context, competitive with Claude Sonnet 4.5 on coding benchmarks.
Pricing
Launch price (from Sep 30, 2025)
$0.600 in / $2.20 out per 1M tokens
Current price (since Sep 30, 2025)
$0.600 in / $2.20 out per 1M tokens
| Tier | Effective | In $/M | Out $/M | Cache $/M | Per image | Per second | Source |
|---|---|---|---|---|---|---|---|
| Standard | Sep 30, 2025 – now | $0.600 | $2.20 | - | - | - | - |
Specifications
| Model type | Text LLM |
|---|---|
| Architecture | MoE |
| Parameters | 357 (32 active) |
| Context window | 200K tokens |
| Max output | 128K tokens |
| Knowledge cutoff | Not disclosed |
| License | MIT License |
| Modalities | Text input Text output |
| Tags | Agentic Coding Reasoning Tool use |
| End of life | Not disclosed |
| Weights | Download |
| Last verified | Not disclosed |