Skip to main content

Video

POST /v1/videos/generations generates video from a text prompt. Billed per second of video.

Example request

curl https://octalabs.id/v1/videos/generations \  -H "Authorization: Bearer $API_KEY" \  -H "Content-Type: application/json" \  -d '{    "model": "VIDEO_MODEL_ID",    "prompt": "Gentle waves on Kuta beach at sunset",    "duration_seconds": 5  }'

Fields

Field Notes
modelRequired. A video model ID from the Models & Pricing page (Video tab).
promptRequired.
duration_secondsRequired. A positive number, at most 3600. It determines how much is held.
nOptional, 1–10 (default 1).
size, fps, userOptional, passed to the model if it supports them.

Cost

The price is per second and metered per millisecond, so a 5.5-second video costs 5.5 × the per-second price; it is not rounded up to whole seconds. The final charge uses the duration the provider confirms, or the duration you requested if the provider doesn’t report one. You are never charged for more than the duration you requested.

The response has to arrive within 90 seconds. A generation that takes longer is answered with 422 timeout_error, and what the provider still delivers is charged, marked Partial. Keep duration_seconds and n small enough to finish in time, and check the Usage page before sending the request again.

Start with any balance,
however small.

Create a free accountSign in AI API · pay as you go