GLM-5 targets coding and agent scenarios. Built on a 744B foundation with asynchronous reinforcement learning and sparse attention, it supports complex systems engineering and long-horizon tasks.
Inference Service Provider
The inference service provider for glm-5 is QwenCloud.
Model Capabilities
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text | Output Modality | Text |
Model Experience | Supported | Function Calling | Supported |
Structured Outputs | Unsupported | Web Search | Unsupported |
Prefix Completion | Unsupported | Context Caching | Supported |
Batch Inference | Unsupported | Fine-tuning | Unsupported |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 169984 | Max Output Length | 16384 |
Context Window | 202752 | Max Input Length (Thinking Mode) | 169984 |
Max Output Length (Thinking Mode) | 16384 | Max Chain-of-Thought Length | 32768 |
Pricing
This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit QwenCloud console for promotional offers.
- China North 2 (Beijing)
Input<=32k
Input<=32k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.573 | Per 1M tokens |
Output | 2.58 | Per 1M tokens |
Input(Implicit Cache) | 0.115 | Per 1M tokens |
32k<Input<=200k
32k<Input<=200k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.86 | Per 1M tokens |
Output | 3.154 | Per 1M tokens |
Input(Implicit Cache) | 0.172 | Per 1M tokens |
Rate Limits
- China North 2 (Beijing)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 500 |
TPM (Tokens Per Minute) | 1,000,000 |