Z.aiGLM-4 Flash
GLM-4.7-Flash
A smaller mixture-of-experts GLM release for local deployment.
Practical use
Practical guidance. Compare it with a larger model on the same tool-assisted tasks. Track missed requirements and incorrect edits separately from latency, then decide which task types justify using the larger model.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API, Downloadable |
| Model identifiers | glm-4.7-flashglm-4.7-flashxzai-org/GLM-4.7-Flash |
| Size | 30-billion-parameter class, with three billion active parameters.GLM-4.7-Flash model card |
| Licence | MIT.GLM-4.7-Flash model card |
| API variants | GLM-4.7-Flash and GLM-4.7-FlashX.Current Z.ai API catalogue |