---
title: rate_limit.exceeded
description: >-
  One IP address or one API key sent more requests than the limit allows in a
  minute; wait and retry.
summary: >-
  Returned with 429 when an IP address passes 1,200 requests a minute or an API
  key passes 600. Some routes, such as sign-in, have a lower limit of their own.
date: 2026-11-11T00:00:00.000Z
type: error
---


`rate_limit.exceeded` is returned when one IP address sends more than 1,200 requests in a minute, or one API key sends more than 600. Sign-in, token and registration routes allow 10 to 20 requests a minute per IP.

## Cause

A client sent more requests than the limit allows inside the current one-minute window. Office networks and shared proxies can reach the IP limit because every user behind them shares one address.

## Resolution

Wait the number of seconds in the `Retry-After` header and send the request again. The `RateLimit-Limit`, `RateLimit-Remaining` and `RateLimit-Reset` headers show the limit and when the window resets. A refused request places no call and debits nothing. Calls that would exceed 10 requests in flight for a workspace use [`ratelimit.concurrent_cap_reached`](/developers/errors/ratelimit.concurrent_cap_reached) instead.

## Example response

```json
{
  "error": {
    "code": "rate_limit.exceeded",
    "message": "too many requests — please slow down and try again later",
    "request_id": "req_01900000abc"
  }
}
```

## Next steps

- See [Rate limits](/developers/concepts/rate-limits) for every limit and the routes it covers.