Skip to content

Rate limits

Gapwise does not currently publish a fixed per-minute or per-day request quota for the public v1 API.

The API runs on managed serverless infrastructure and is protected by platform-level abuse and availability controls. Those controls can change independently of the API contract, so documentation does not invent an exact global number that the service cannot guarantee.

  • Normal interactive and development usage should keep requests bounded and cache reusable public data.
  • A request can receive 429 Too Many Requests when platform protection or future service controls require it.
  • Transient 5xx responses are also possible during infrastructure or upstream failures.
  • Do not build correctness around an assumed hidden quota.

For 429 or transient 5xx responses, retry only idempotent requests automatically. Use exponential backoff with jitter and a low retry ceiling. For POST route or gap calculations, only retry when your application can safely repeat the same deterministic request.

A reasonable client pattern is:

  1. wait roughly 0.5–1 second after the first transient failure;
  2. increase the delay exponentially;
  3. add random jitter so many clients do not retry at the same instant;
  4. stop after a small number of attempts and surface the failure.

Never retry 400, 404, 405, 409, 413, or 415 errors without changing the request.

Building identity and provenance change slowly and may be cached by the service. Place availability is more time-sensitive. Route and gap calculations should be cached by your application only when the inputs and data-version assumptions make reuse correct.

If you need sustained high-volume access, prefer the versioned public campus snapshot for data that does not require a calculation, and avoid repeatedly polling unchanged resources. The public API is designed as a developer surface, not as an unrestricted scraping endpoint.