Skip to main content

Base URL

Any other path under /v1, such as /v1/embeddings or /v1/models/{id}, returns 404 with an empty body. A method the path does not accept, such as GET /v1/chat/completions, returns 405 with an empty body.

Authentication

Content-Type is required on requests with a body. Keys come from the Fortytwo platform — see Get access. Cross-origin requests from a browser are not supported.

Model IDs

GET /v1/models returns model IDs only. Limits and features of each model are on Models, prices on Pricing. A request that uses a feature the model does not support, such as reasoning, tools, structured output or images, fails with an error before generation.

AI-generated output

Every chat completion that carries generated output, streamed or not, has the header x-ai-generated: true. Errors and GET /v1/models do not.

Request IDs

Every response has an x-request-id header. Error bodies repeat it as request_id. Log this ID: support uses it to find a request. A value sent by the client is ignored.

Errors

Errors from both endpoints use one JSON envelope. Statuses, codes and retry rules are on Errors.
Last modified on October 22, 2026