Skip to main content
All API endpoints include performance expectations using this format: p50 <= X ms, p99 <= Y ms This page explains how to interpret these values.

​
Percentiles

  • p50 is the median response time. 50% of successful requests complete at or below this value.
  • p99 is the tail response time. 99% of successful requests complete at or below this value.
Latency scope is server-side successful request handling only (2xx), excluding network/client effects.