Continue

One /v1 endpoint for every provider. Point your OpenAI client at a new base URL and key, bring your own provider keys, and pay 0% markup on tokens.

Continue reads its models from config.yaml, at ~/.continue/config.yaml on macOS and Linux and %USERPROFILE%\.continue\config.yaml on Windows. Add an entry with the openai provider and our base URL:

yaml
name: My Config
version: 0.0.1
schema: v1
models:
  - name: Claude Sonnet 5 (Autonomous Relay)
    provider: openai
    model: anthropic/claude-sonnet-5
    apiBase: https://autonomousrelay.com/v1
    apiKey: ${{ secrets.AUTONOMOUSRELAY_API_KEY }}
    roles: [chat, edit, apply]
    capabilities: [tool_use]
  • capabilities: [tool_use] turns on tool calling, which Continue’s agent mode needs. Leave it out for a model without tool support.
  • Put the key in a .env file Continue reads: the workspace’s .env, .continue/.env or ~/.continue/.env, as AUTONOMOUSRELAY_API_KEY=…. An IDE extension cannot see a variable you exported in a shell. Restart the IDE after changing it.
  • Leave useLegacyCompletionsEndpoint unset: the model is served on chat completions.