Rate limits need to be something you can check quickly if they are to be of much use. We take performance very seriously and we consistently deliver fast responses through a combination of HTTP2 via GRPC to reuse connections, NLBs in AWS, and efficient code on our side.
But sometimes a picture is worth a thousand words.
Actual response times from the rate limit API for the past week:
These are median, 75th, 98th and 99th percentiles for server side latency for acquiring a rate limit from Prefab.cloud. End user times will vary primarily based on latency to AWS US East-1. Average client response times are typically < 20ms.Try it