> ## Documentation Index
> Fetch the complete documentation index at: https://developers.artport.co/llms.txt
> Use this file to discover all available pages before exploring further.

# Rate Limiting

> Explore the limitations and guidelines

### Overview

In order to ensure that the API remains stable for all customers, we enforce a limit on the number of requests that can be sent.

<Note>
  The API limit is currently 20 requests per minute.
</Note>

### Exceeding the limit

If you send more than the permitted number of requests per minute, then those requests will be rejected. Until the limit resets, all requests will receive a `429 Too Many Requests` response.

### Managing rate limits

Usually, a `429 Too Many Requests` response isn't an issue. You can just send it again in a minute or so. However, if you receive this error regularly, then you should adjust your integration so that it backs off.

### Developing a backoff strategy

When you receive a response, it will include the following headers that indicate your current usage of the API:

| Header                | Description                                          |
| --------------------- | ---------------------------------------------------- |
| X-RateLimit-Limit     | The number of requests that can be sent per minute.  |
| X-RateLimit-Remaining | The number of requests that can be sent this minute. |

If you exceed the API rate limit, then the following headers will also be present in the response:

| Header            | Description                                       |
| ----------------- | ------------------------------------------------- |
| Retry-After       | Seconds remaining until new requests can be sent. |
| X-RateLimit-Reset | The unix timestamp when new requests can be sent. |

#### Tips

You can use `X-RateLimit-Remaining` to avoid hitting the rate limit. If you don't have any requests available, then there is no point sending one as it will return a `429 Too Many Requests` response.

If you hit the rate limit / receive a `429 Too Many Requests` response, then you can use `Retry-After` or `X-RateLimit-Reset` to determine when it is suitable to send additional requests to the API.