Rate limits
No rate limits, free credits included, the one request we ask you to keep an eye on, and what a 429 or 503 means.
No rate limits, free credits included. We ask that you stay below 50 requests in progress at once.
"In progress" means requests we have accepted and not yet answered, plus background jobs still queued or running. It is counted per team, not per key. Send as many requests as you like over time; just do not start a large number all at the same moment. A batch counts each item.
Too many at once
Past a very large number of requests in progress, new ones are refused with 503 overloaded. It is retryable and free: nothing is charged, and a retry-after header says how many seconds to wait.
{
"ok": false,
"error": { "code": "overloaded", "message": "This account has too many requests in progress at once; retry shortly. You are not charged for this response.", "retryable": true },
"credits_charged": 0,
"request_id": "req_0123456789abcdef01234567"
}
Wait for retry-after, then continue. The retry helper already does this. Repeats of something you already fetched (see caching) are answered from the cache and do not count.
A flood from one address
Far past anything a normal script sends, requests from one address are refused with 429 rate_limited. It is retryable and free, with a retry-after header holding the number of seconds to wait. No account has a per-minute limit.
The platforms have limits too
When Instagram, TikTok or another platform throttles us, the answer is 503 upstream_rate_limited. It is retryable and free. That is the platform's limit, not yours.
More capacity
If you need to run far more than 50 at once, write to [email protected] and we will raise it for your team.