Skip to content
Inferray
DashboardGet API key

Search guides, API endpoints, integrations and coding agents.

Docs · GitHub Copilot

GitHub Copilot

GitHub Copilot lets you bring your own model provider for chat and agent sessions. Add Inferray as an OpenAI-compatible provider to use any model in the catalogue.

GitHub Copilot app

  1. Sign in with your GitHub account.
  2. Open Settings → Model providers and click Add provider.
  3. Choose OpenAI-compatible.
  4. Enter a name, your Inferray key, and the base URL https://api.inferray.com/v1.
  5. Save, then pick an Inferray model in the session's model picker.

VS Code

In Copilot Chat, open the model picker and choose Manage Models…. Where your version offers an OpenAI Compatible provider, add it with the base URL https://api.inferray.com/v1 and your Inferray key, then add the model IDs you want from the catalogue.

Mark models that support tools as tool-capable so agent mode can use them, and set their context and output limits from the catalogue.

What uses your key

Chat and agent requests with an Inferray model are billed to your Inferray project. Copilot's inline completions — the grey suggestions you accept with Tab — keep using GitHub's own models; they do not support custom providers. GitHub sign-in is still required, and organisation policies can restrict custom providers on managed accounts.

Troubleshooting

  • The provider or model is missing — update the app or VS Code, and check the saved base URL and key.
  • Agent tasks fail — choose a model with tool calling and streaming support.

Models that work well here