# Rate limits (/rate-limits)

Each API key can make **120 requests per minute**. Keys don't share the limit, so two systems with their own keys don't slow each other down. The count starts over each minute.

Over the limit, the API answers `429` with a `Retry-After` header, in seconds:

```http
HTTP/1.1 429 Too Many Requests
Retry-After: 42
Content-Type: application/json

{ "error": { "code": "rate_limited", "message": "Too many API requests: the limit is 120 per minute for this key. Try again in 42 seconds." } }
```

## Adding and removing leads [#adding-and-removing-leads]

[Add leads](/api-reference/leads/addLeads) and [Remove leads](/api-reference/leads/removeLeads) have a second limit that all keys of a workspace share: **120 calls per minute and 5,000 calls per day**. Each call can carry up to 500 people, so this is rarely reached. It answers with the same `429`.

## Staying under the limit [#staying-under-the-limit]

- Wait for `Retry-After` before trying again.
- Send leads in batches: up to 500 people per [Add leads](/api-reference/leads/addLeads) call, not one call per person.
- Use `limit=100` when reading lists.

## The MCP server [#the-mcp-server]

The [Alienmail MCP server](/guides/mcp-server) has its own limit: 60 tool calls per minute and 2,000 per day for each API key, and the same for each app connected by signing in. The limits on its sign-in endpoints are in [Connect your own app](/guides/mcp-oauth#limits).