n8n

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.

n8n’s OpenAI credential and its OpenAI Chat Model node both take a base URL, so workflows and AI agents in n8n can use any model in the catalogue.

The credential

  • In Credentials, create an OpenAI credential.
  • API Key: an inference key from Keys in the dashboard.
  • Base URL: https://autonomousrelay.com/v1. n8n tests the credential by listing /models, which answers with the catalogue.

The chat model node

  • Add an OpenAI Chat Model node with that credential.
  • For Model, pick from the list or choose By ID and type the catalogue id, such as anthropic/claude-sonnet-5.
  • Turn Use Responses API off. In recent versions of the node it is on by default, and chat completions is the endpoint every model here is served on.
n8n warns that models reached through a base URL override may not all support tool calling or JSON output. Check the model’s page in the catalogue before giving it tools in an agent.