Docs · API
Rate limits and fair use
Per-token limits
600 requests per minute per token, burstable to 1200 for up to 10 seconds. Above the burst ceiling, requests receive HTTP 429 with a Retry-After header.
Per-endpoint limits
Write endpoints are more tightly controlled than read endpoints. See the reference table in the workspace docs section.
Fair use
The published rate limits are generous for legitimate operational use. Requests obviously abusing the API for high-frequency scraping are subject to the Acceptable Use Policy and can be throttled or blocked.