GitHub Copilot Setup Guide
Use UnoRouter models in VS Code Copilot Chat and agent mode through a Custom Endpoint.
Overview
VS Code lets Copilot Chat use models from your own key. The Custom Endpoint provider accepts any OpenAI-compatible Chat Completions URL and works without a Copilot plan.
Quick Config
Paste these into the client. Your key auto-fills when signed in.
Base URL: https://api.unorouter.com/v1/chat/completions
API Key: YOUR_API_KEYCompatibility
Step-by-step setup
- 1Open Manage Language Models
Run Chat: Manage Language Models from the Command Palette, or open the model picker in Chat and click the gear icon.
- 2Add a Custom Endpoint
Click Add Models and choose Custom Endpoint. Enter a group name, a display name and your UnoRouter key, and keep the API type on Chat Completions.
- 3Edit the model list
VS Code opens chatLanguageModels.json. Put these models into the group it created, with the full chat completions URL for each. Keep toolCalling on for agent mode.
json"models": [ { "id": "glm-5.3-flash:free", "name": "GLM 5.3 Flash (free)", "url": "https://api.unorouter.com/v1/chat/completions", "toolCalling": true, "vision": false, "maxInputTokens": 128000, "maxOutputTokens": 16000 }, { "id": "claude-sonnet-5-5", "name": "Claude Sonnet 5.5", "url": "https://api.unorouter.com/v1/chat/completions", "toolCalling": true, "vision": true, "maxInputTokens": 200000, "maxOutputTokens": 32000 } ] - 4Pick the model
Choose the model in the Chat model picker. Restart VS Code if it does not show up.
Best free models for GitHub Copilot
Free models that work here. Paste any model id into the client.
Gotchas
- Your own models cover chat, agent mode and tools. Inline code completions and next edit suggestions still need a GitHub sign-in and Copilot.
- On Copilot Business and Enterprise, an admin may need to enable the Bring Your Own Language Model Key policy first.
- The older OpenAI Compatible provider and the customOAIModels setting are deprecated. Use Custom Endpoint instead.