Model family
GLM Models
GLM models from Z.ai are used for agentic engineering, tool use, coding, and reasoning workflows.
Best for
Agents
Use this family hub to compare GLM variants for agents workflows, then open the detail page for deeper deployment notes.
Code
Use this family hub to compare GLM variants for code workflows, then open the detail page for deeper deployment notes.
Reasoning
Use this family hub to compare GLM variants for reasoning workflows, then open the detail page for deeper deployment notes.
Source box
This family currently includes 2 records tied to an exact published checkpoint. Identity is recorded per model so a representative checkpoint is never treated as the whole family.
Identity checked: 2026-08-21
Artifact identity does not establish licence or context truth. Those checks remain separate.
Variants
GLM models grouped by workflow
Latest / flagship
Coding
Compare
All GLM models in the directory
| Model | Type | Best for | Local runner notes | License | Detail |
|---|---|---|---|---|---|
| GLM-5.1 | Agents | Teams evaluating open-weight models for tool-calling, software engineering, and agent systems. | Full model is server-class; smaller Air or Flash style variants are usually better for local testing. | MIT | Open |
| GLM-4.5 | Agents | Teams comparing prior GLM releases with current GLM-5.1 behavior. | Full model is server-class; Air or quantized builds are more practical. | MIT | Open |
| GLM-5.2 | Code | GLM-5.2 is Z.ai's large mixture-of-experts model, aimed at coding, agentic tool-use, and very long context work rather than single-GPU home setups. | GLM-5.2 is a 753B MoE model requiring multi-GPU or distributed inference for full deployment. Ollama publishes a cloud-hosted entry only (`glm-5.2:cloud`, re-checked 2026-08-24) and exposes no downloadable local artifact, so this catalog records no local tag. Below data-center scale, the unsloth/GLM-5.2-GGUF 1-bit UD-IQ1 variants (~217–228 GB) via llama.cpp or Unsloth Studio with KTransformers are the only local execution path, at reduced fidelity. Not consumer-local hardware friendly. | MIT | Open |