<div class="section-label"><span class="num">05</span> Throttle Limits</div> <div class="section-title">Rate limits keep the API <em>reliable</em>.</div> If your application sends many requests in a short window, the API may respond with **429 Too Many Requests**. That means you have hit the service rate limit. Back off and retry later. Persistent 429s usually mean the client is posting faster than the account is allowed, not that the prediction itself failed. A 429 is distinct from [[API/Getting Started/Payload Limits|413 Payload Too Large]] (body too big) and from a completed job whose results you retrieve with [[API/Prediction Engine/Results|GET /results]]. ## Related - [[API/Introduction|API]] - [[API/Getting Started/Payload Limits|Payload Limits]] · [[API/Prediction Engine/Results|Results]]