Prefect Cloud has controls in place to ensure stability.

​
API Rate Limits

Prefect Cloud rate limits are subject to changeAPI rate limits are subject to change. Contact Prefect support at help@prefect.io with questions about rate limits.
Prefect Cloud’s API rate limits restrict the number of requests that a single client can make to certain endpoints in a given time period. The flow_runs, task_runs, and flows endpoints and their subroutes are limited to:
  • 400 requests per minute for Free accounts
  • 2,000 requests per minute for Pro accounts
These endpoints return a 429 response with an appropriate Retry-After header if this limit is triggered. See ClientSettings for more information on how retries are handled client-side and how to modify the default behavior. The logs endpoint is limited to:
  • 700 logs per minute for Free accounts
  • 10,000 logs per minute for Pro accounts
The endpoint returns a 429 response if this limit is triggered.

​
Metadata retention period

Flow run, task run, and artifact metadata is retained according to your Prefect Cloud account’s retention period. The retention period applies to all workspaces that belong to the account. The retention period is the number of days that metadata is available after it is created. For flow and task runs, it is calculated from the time the run reaches a terminal state. Subflow runs are retained independently from their parent flow runs. They are removed based on the time each subflow run reaches a terminal state. If you have needs that require a custom retention period, contact Prefect’s sales team.