Model maker
Z.ai develops the GLM family of language models. Borg offers a selection of GLM generations for direct use and routing.
Search by model name, lane id or use case.
| Model | Context | Reasoning | Release | Best for | Borg speed | Bench |
|---|---|---|---|---|---|---|
| GLM-4.5 | 131,072 tokens Vendor source | Yes | 2025-07-28 Vendor source | tool-assisted software engineering | — | — |
| GLM-4.5-Air | 131,072 tokens Vendor source | Yes | 2025-07-28 Vendor source | cost-conscious coding-agent steps | 12.1 output tokens/s (whole turn) | — |
| GLM-4.6 | 200K tokens (vendor-stated) Vendor source | Yes | 2025-09-30 Vendor source | large-repository code review | — | — |
| GLM-4.7 | 200K tokens (vendor-stated) Vendor source | Yes | 2025-12-22 Vendor source | end-to-end application prototypes | 40.6 output tokens/s (whole turn) | — |
| GLM-5 | 200K tokens (vendor-stated) Vendor source | Yes | 2026-02-12 Vendor source | multi-stage systems engineering | 33.3 output tokens/s (whole turn) | — |
| GLM-5-Turbo | 200K tokens (vendor-stated) Vendor source | Yes | — | OpenClaw agent automation | 16.4 output tokens/s (whole turn) | 60% |
| GLM-5.1 | 200K tokens (vendor-stated) Vendor source | Yes | 2026-04-07 Vendor source | long-running autonomous coding | — | — |
| GLM-5.2 | 1,048,576 tokens Vendor source | Yes | 2026-06-16 Vendor source | whole-repository architecture audits | 25.7 output tokens/s (whole turn) | 47% |
| GLM-5.3 | 1,048,576 tokens Vendor source | Yes | 2026-08-18 Vendor source | complex coding-agent tasks | 25.2 output tokens/s (whole turn) | 53% |
These are the Z.ai models currently listed as live in the Borg directory. The descriptions below summarize each model's published use cases; the linked model pages give the full account. A lane id is the value you use to select a model directly in the Borg gateway, not a separate model name. Check the linked pages for context windows, reasoning support and any vendor API list prices before choosing a lane for your work.
GLM-4.5 is the larger member of Z.ai's original agent-focused GLM-4.5 pair, with a mixture-of-experts design and more active capacity than GLM-4.5-Air.
Best for GLM-4.5:
Borg lane ids: glm-4.5, z-ai/glm-4.5. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-4.5-Air is the leaner member of the first GLM-4.5 agent-oriented pair.
Best for GLM-4.5-Air:
Borg lane ids: glm-4.5-air, z-ai/glm-4.5-air. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-4.6 follows GLM-4.5 as Z.ai's text-focused coding and agent model.
Best for GLM-4.6:
Borg lane ids: glm-4.6, z-ai/glm-4.6. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-4.7 is the successor to GLM-4.6 in Z.ai's text-only GLM line.
Best for GLM-4.7:
Borg lane ids: glm-4.7, z-ai/glm-4.7. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-5 begins Z.ai's next foundation-model generation after GLM-4.7, with a focus on agentic engineering rather than just coding completion.
Best for GLM-5:
Borg lane ids: glm-5, z-ai/glm-5. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-5-Turbo is a public Z.ai variant optimized around OpenClaw-style agent workflows, rather than a generic promise of faster responses.
Best for GLM-5-Turbo:
Borg lane ids: glm-5-turbo, z-ai/glm-5-turbo. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-5.1 extends the GLM-5 foundation with a specific emphasis on sustained autonomous work.
Best for GLM-5.1:
Borg lane ids: glm-5.1, z-ai/glm-5.1. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-5.2 is the long-context successor to GLM-5.1, retaining a focus on sustained engineering but substantially expanding how much project material one request can hold.
Best for GLM-5.2:
Borg lane ids: glm-5.2, z-ai/glm-5.2. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
GLM-5.3 builds on the same base model as GLM-5.2, with improvements from post-training rather than a new base architecture.
Best for GLM-5.3:
Borg lane ids: glm-5.3, z-ai/glm-5.3. Open the model page for its full description, available specifications, vendor price sources, and routing roles.
Borg Fusion can route between models in the subscription catalog according to the task and the selected profile. Having a Z.ai lane in the catalog does not mean that every Fusion profile will choose it for every request. No specific routing assignments for Z.ai are published in this catalog. The model pages distinguish published routing roles from use cases, so you can compare options without assuming a benchmark result or a routing guarantee. You can also choose a listed lane directly if you want to control which model handles a request.
The table and model details above list the current live Z.ai models in Borg. Each linked page includes the model's lane ids and available specifications. Models awaiting identity review are not shown as live.
Fusion profiles select from available lanes depending on the task. Consult each Z.ai model page for its published Fusion roles; a missing role is not evidence that the model cannot be selected directly.
Yes. Open a Z.ai model page to find its Borg lane ids and the compatible gateway model field. Use the lane id rather than the public page slug when choosing a model directly.
Compare Z.ai with other makers in the same Borg subscription: Anthropic · OpenAI · Google · xAI · Moonshot AI · Meta · Borg
Z.ai
Anthropic
OpenAI
Google
xAI
Moonshot AI
Meta