# Rate limits

> Each API key can make 60 requests a minute, and each IP address 120. Every response carries headers that say how much of the limit is left.

- Audience: API
- Page type: reference
- Updated: 2026-10-08
- Canonical URL: https://www.rendezvu.co/docs/api/rate-limits

The Brand API limits how fast each key and each IP address can call it, so one busy integration cannot slow the API down for everyone else.

## Limits

| Limit | Requests per minute |
| --- | --- |
| Per API key | 60 |
| Per IP address | 120 |

The IP limit is checked first, before the key, so it also covers requests with a bad key. Both windows are one minute long.

## Rate limit headers

Responses carry the state of the limit:

| Header | Meaning |
| --- | --- |
| `X-RateLimit-Limit` | Requests allowed in the window. |
| `X-RateLimit-Remaining` | Requests left in the window. |
| `X-RateLimit-Reset` | When the window resets, as an ISO 8601 timestamp. |
| `Retry-After` | On a `429` only: seconds to wait before trying again. |

On an authenticated request the headers describe your key's limit.

## Stay under the limit

- Page with `per_page=100`. It is the same data in a quarter of the requests.
- Sync on a schedule rather than on demand, and store what you read.
- Run one sync at a time per key. Parallel workers on one key share its 60 requests.
- On a `429`, wait for `Retry-After` and then back off exponentially. See [Errors](/docs/api/errors) for a retry loop.

## Related

- [Errors](https://www.rendezvu.co/docs/api/errors.md): The Brand API uses conventional HTTP status codes. 2xx means success, 4xx means something in the request needs to change, and 5xx means retry later.
- [Pagination and requests](https://www.rendezvu.co/docs/api/pagination.md): Every response shares one JSON envelope. List endpoints page with page and per_page, query parameters are strict, and every date is ISO 8601 in UTC.

---

Source: https://www.rendezvu.co/docs/api/rate-limits (Rendezvu docs, Markdown view). Every page: https://www.rendezvu.co/llms.txt