Skip to main content
Most api.telnyx.com endpoints enforce rate limits. Limits and scopes can differ by product and operation and can change as service capacity changes. Treat response headers and product documentation as authoritative rather than hard-coding a global request rate. Limits are counted per region. Requests made against one regional endpoint (for example, api.telnyx.eu) do not count toward the limits of another (for example, api.telnyx.com) — a 429 Too Many Requests on one region’s endpoint does not indicate the state of another region’s limits. See Regional endpoints & the EU PoP.

​
Rate-limit headers

When present, these headers describe the current request window: Header availability and semantics can vary by endpoint. A missing header is not evidence that an endpoint has no limit. Follow the selected product’s documentation before interpreting x-ratelimit-reset as either a duration or a timestamp.

​
429 response

A throttled REST request returns HTTP 429 Too Many Requests:

​
Recover safely

  1. Stop issuing requests in the affected scope.
  2. Honor Retry-After when present; otherwise use the documented reset signal or bounded exponential backoff with jitter.
  3. Retry only when the original operation is safe to repeat.
  4. Reduce concurrency and cache safe read results where appropriate.
  5. Bound attempts and total elapsed retry time.
Do not distribute traffic across API keys to bypass a product limit. Contact Telnyx Support when normal operation consistently exceeds documented capacity.

​
Product-specific limits

Other products publish limits in their product guides or endpoint descriptions. Do not infer Messaging limits for Voice, Storage, Wireless, or other APIs.