> ## Documentation Index
> Fetch the complete documentation index at: https://docs-agentic-id.icustomer.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Rate limits

> The request limits per caller and per address, and how to handle a 429.

Three limits apply. Each counts requests in a fixed one-minute window.

| Limit | Counted per | By default | Applies to |
| - | - | - | - |
| Caller | API key, or signed-in account | 600 requests a minute | Every authenticated request |
| Address | IP address | 1,800 requests a minute | Every request, before the credential is checked |
| Refused credentials | IP address | 20 a minute | Counts refused credentials. Once reached, holds back requests that have no valid credential. |

The numbers are defaults. Your deployment may be set differently.

## How each limit behaves

**Caller.** Each API key has its own window. A signed-in account has its own window across its sessions.

**Address.** A ceiling above the caller limit. Many keys behind one address share it.

**Refused credentials.** After the limit of refused credentials from one address in a minute, further requests from that address without a valid credential get `429` until the window ends. Requests with a valid credential from the same address still pass.

## The 429 response

```http theme={null}
HTTP/1.1 429 Too Many Requests
Content-Type: application/problem+json
Retry-After: 23
X-Request-Id: req_6f0e1c1e-3a1b-4c55-9a0e-2f7b9d1c8a11
```

```json theme={null}
{
  "type": "https://docs.agentic-id.icustomer.ai/errors/rate_limited",
  "title": "Too many requests",
  "status": 429,
  "code": "rate_limited",
  "detail": "Too many requests for this caller; retry in 23s.",
  "retryAfterSeconds": 23,
  "request_id": "req_6f0e1c1e-3a1b-4c55-9a0e-2f7b9d1c8a11"
}
```

| Field | Meaning |
| - | - |
| `Retry-After` header | Seconds until the current window ends |
| `retryAfterSeconds` | The same number, in the body |
| `detail` | Which limit was hit: this caller, this address, or refused credentials from this address |

## Handle a 429

* Wait for `Retry-After` seconds, then send the same request.
* A `429` is refused before any work. Nothing was charged, and a retry cannot charge twice.
* Send records in batches: up to 1,000 emails in one validate call and up to 100 contacts in one enrich call. One batch counts as one request.
* The [MCP server](https://docs-agentic-id.icustomer.ai/mcp-server.md) and the [CLI](https://docs-agentic-id.icustomer.ai/cli.md) wait and retry a `429` once for you.

There are no rate-limit headers on successful responses.

Related: [Errors](https://docs-agentic-id.icustomer.ai/errors.md), [Idempotency](https://docs-agentic-id.icustomer.ai/idempotency.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.