Skip to main content
To keep the API fast and reliable for everyone, requests are rate limited per API key. These limits apply globally across all endpoints in Conductor’s API.

​
Global rate limits

​
Budget by call, not by workflow step

When estimating your usage, count individual API calls—not the number of nodes in a workflow. A single node can generate many calls depending on the data you are querying. This matters most for asynchronous endpoints: polling an async query for its results will usually require several calls before the data is ready, and each poll counts against both your per-second and per-day limits.
When working with async endpoints, use a reasonable polling interval so you don’t exhaust your per-second limit while waiting for results. See Async Queries for the recommended pattern.