Rate limits and retries

Respect shared fixed-window workspace limits and Retry-After responses.

Limits are workspace-wide across all public endpoints:

WindowLimit
Second10 requests
Minute60 requests
UTC day5000 requests

Windows are fixed. Rejected request returns:

HTTP/1.1 429 Too Many Requests
Retry-After: 1

Wait at least Retry-After seconds. For repeated transient failures, use exponential backoff with jitter.

All schedulers, workers, ReadMe Try It requests, JSON requests, and recording requests share same workspace quota. Coordinate concurrency centrally.


Did this page help you?