Getting started
Rate limits
The request-rate ceiling and the headers that track it.
How it works
In addition to your monthly quota, a request-rate limit applies to each API key. The default ceiling is 100 requests per 60-second window, enforced in a sliding window per key.
When you exceed it you receive 429 Too Many Requests:
{
"error": {
"code": "RATE_LIMIT_EXCEEDED",
"message": "Rate limit exceeded. Limit: 100 requests per 60 seconds"
}
}There is no Retry-After header — use the X-RateLimit-* headers below.
Response headers
Responses carry three headers:
| Header | Value |
|---|---|
X-RateLimit-Limit | The ceiling for the throttle that stamped this response |
X-RateLimit-Remaining | How much of it is left |
X-RateLimit-Reset | When it resets |
Use X-RateLimit-Remaining to implement client-side back-pressure before hitting the limit.