What is API rate limiting?

Definition

API rate limiting protects a service from overload, abuse, and unfair resource consumption by controlling request volume. Limits may apply per account, credential, address, model, or rolling time window.

When a client exceeds the limit, services commonly return HTTP status 429 and may provide retry guidance. Long-running applications need backoff, queueing, and durable progress so temporary limits do not corrupt or abandon work.

ELI5

API rate limiting controls how many requests a client can send to a service during a period of time. It protects the service from overload and prevents one client from using an unfair share of the available capacity.

For example, an API might allow 100 requests per minute and temporarily reject the next request when that allowance is used. A well-behaved application waits according to the service guidance, keeps its progress and avoids sending the same consequential operation twice.

Frequently asked questions

What does HTTP 429 mean?

It means the client has sent more requests than the service currently permits under its applicable rate-limit policy.

How should an application handle rate limits?

It should respect retry guidance, use bounded backoff, control concurrency, queue work, and preserve progress across interruptions.

Videos explaining API rate limiting