LibreChat

LibreChat supports custom OpenAI-compatible endpoints. Add Inferray in librechat.yaml:

endpoints:
  custom:
    - name: "Inferray"
      apiKey: "${INFERRAY_API_KEY}"
      baseURL: "https://api.inferray.com/v1"
      models:
        default: ["gpt-5.5", "claude-sonnet-4-6", "gemini-3.6-flash"]
        fetch: true
      titleConvo: true
      titleModel: "gemini-3.6-flash"
      modelDisplayLabel: "Inferray"

Put the key in LibreChat's .env:

INFERRAY_API_KEY=inferray_...

With fetch: true, LibreChat loads the full model list from GET /v1/models; default is what it shows if that fails. A small, fast model makes a good titleModel, since it runs on every new conversation.

Restart LibreChat after editing the file, then pick Inferray in the endpoint menu.