Rate limits

When a rate-limited operation returns 429 Too Many Requests, it also returns a Retry-After header.

Retry-After: 30

Wait for the specified number of seconds before retrying. Use exponential backoff with jitter for repeated failures and avoid parallel retries for the same resource.

Rate-limit behavior is endpoint-specific. Do not assume that an endpoint returns 429 when the API Reference does not document it. Public RAG is limited per API key and documents 429 with Retry-After on its search endpoint.