Usage limits
Rate limits are per key (unless noted). When you exceed them, the API returns 429.
| What | Limit |
|---|---|
| Reads | 120/min |
| Catalog writes | 60/min |
| Generation (sync + batch submit) | 30/min |
| Export | 10/min |
| Products per upsert | 500 |
| Products per batch | 500 |
| Body size | 2 MB |
| Active keys per tenant | 10 |
Exceeding a rate limit returns 429 with a Retry-After header (in seconds).
Charges use the same credits and prices as the app (no surcharge). Your daily AI spend cap is shared with usage from the app: on reaching it, generation also returns 429, but without Retry-After (it resets the next day, in UTC).