Rate limits

Every API key gets 600 requests per minute, shared across /v1/api/* and the MCP endpoint. Requests without a valid key are limited per IP address instead.

#Headers

Every response, success or error, carries:

When you exceed the limit you get 429 with code resource_exhausted and a Retry-After header:

HTTP/1.1 429 Too Many Requests
RateLimit-Limit: 600
RateLimit-Remaining: 0
RateLimit-Reset: 23
Retry-After: 23
Content-Type: application/json

{"error":{"code":"resource_exhausted","message":"Rate limit exceeded (600 requests/minute). Retry after 23s.","requestId":"req_…"}}

The window is a fixed 60 seconds.

#Staying under the limit

• Watch RateLimit-Remaining and slow down as it approaches zero, rather than reacting to 429s.
• Use webhooks instead of polling. Subscribe to message.received and Ody pushes new messages to you. If you must poll a shared inbox, every 10–30 seconds is plenty.
• Request bigger pages (limit up to each endpoint's maximum) when syncing.
• Back off with jitter on 429 and 5xx; honour Retry-After; cap your retries.

async function withBackoff<T>(fn: () => Promise<T>, attempts = 5): Promise<T> {
  for (let i = 0; ; i++) {
    try {
      return await fn();
    } catch (err: any) {
      const retriable = err?.status === 429 || err?.status >= 500;
      if (!retriable || i >= attempts - 1) throw err;
      const wait = err.retryAfter ? err.retryAfter * 1000 : Math.min(30_000, 2 ** i * 500) * (0.5 + Math.random());
      await new Promise((r) => setTimeout(r, wait));
    }
  }
}

Need a higher limit for a specific integration? Email [email protected] with your use case.