New coding models & integrations

Ollama announced on October 16, 2025 that two coding models, GLM-4.6 and Qwen3-coder-480B, are available on its cloud service, with integrations for tools users may already work with. The company also said Qwen3-Coder-30B has been updated for faster, more reliable tool calling in its new engine.

The models can be run via Ollama commands: glm-4.6:cloud and qwen3-coder:480b-cloud for the cloud versions, and qwen3-coder:30b. Ollama stated that qwen3-coder:480b is also available locally for users with more than 300GB of VRAM.

Ollama documented setup instructions for several coding tools. For VS Code, users pull the cloud models, open the Copilot chat sidebar, select Manage models in the model dropdown, choose Ollama under the provider dropdown, pick the desired models, then select a model such as glm-4.6. For Zed, users pull the models, open Zed, click the agent panel button, open the model dropdown and Configure, select Ollama under LLM providers, confirm the host URL is http://localhost:11434 and connect, then select a model. For Droid, users install it via a curl command and add custom model configuration to ~/.factory/config.json specifying glm-4.6:cloud and qwen3-coder:480b-cloud with a base URL of http://localhost:11434/v1.

Ollama's documentation now includes sections on using Ollama with Codex, Cline, VS Code, Zed, Droid and Roo code.

Cloud models such as glm-4.6 and qwen3-coder:480b can also be accessed directly through ollama.com's cloud API. Ollama described creating an API key, setting it in the environment as OLLAMA_API_KEY, and calling the https://ollama.com/api/chat endpoint.