Skip to content

Errors & limits

Make failures understandable and give your application a useful recovery path.

Common responses

StatusMeaningNext step
400The request does not match the supported API.Check the model, message content, and parameters.
401Authentication failed.Check the key and bearer header.
429An account request or concurrency limit was reached.Reduce concurrent work and respect retry guidance.
503Capacity or a required dependency is unavailable.Inspect the error code and Retry-After.

No available provider

no_provider_available means no free matching provider can serve your selected model now. request_capacity means Router’s global request capacity is full. Both return 503.

Input and output limits

The API enforces model-specific limits alongside request size and concurrency limits. Discover the edition’s configured limits before sending long prompts. Unsupported generation parameters are rejected explicitly.

Usage is a measurement

Token counts are provider-reported operational measurements. Missing usage is not zero, and these measurements are not a billing ledger.