Overview
GLM-5-Turbo is a foundation model optimized for OpenClaw and continuous agent workflows. It is designed to decompose complex instructions, plan multi-step tasks, coordinate agents, and use external tools and Skills. (docs.bigmodel.cn)
Capabilities and uses
- Text input and text output; 200K context window and up to 128K output tokens.
- Supports thinking modes, streaming, function calling, context caching, structured output and MCP tool integration.
- Suitable for workflow automation, information gathering, file processing, multi-tool tasks and repeatable business processes. (docs.bigmodel.cn)
Limitations
The documented modalities are text input and output; do not assume the visual capabilities of the separate GLM-5V-Turbo model. The 200K context and 128K maximum output are limits, and agent actions depend on the tools and permissions available in the surrounding application. Generated results and consequential actions should be reviewed. (docs.bigmodel.cn)
Release date
March 15, 2026, according to the official release record. (docs.bigmodel.cn)
API pricing
Mainland China — BigModel, CNY per 1 million tokens, checked November 16, 2026:
| Input length | Input | Output | Cache-hit input |
|---|---|---|---|
| 0 to less than 32K tokens | CNY 5 | CNY 22 | CNY 1.2 |
| 32K tokens or more | CNY 7 | CNY 26 | CNY 1.8 |
Cache storage is listed as free for a limited time. These are pay-as-you-go rates; consult the platform for current terms and actual billing. The official international pricing page was also checked on November 16, 2026, but did not list this model. (docs.bigmodel.cn)