Text-to-Video
POST /runway/pro/generate generates GEN-4 text-to-video with Runway via CometAPI, supporting X-Runway-Version for versioned outputs.
Required fields
callback_urlis requiredprompt,ratio,style, andmodeldefine the generation requestoptions.secondsandoptions.motion_vectorare required
Validated request path
- Header:
X-Runway-Version: 2024-11-06 - Model:
gen4_turbo - Response:
code: 200,msg: success, anddata.task_id
Duration and ratio
| Setting | Supported values | Default starting point | Boundary behavior |
|---|---|---|---|
options.seconds | 5, 10 for standard Runway video generations | 5 | Some compatibility examples use 1 for smoke tests, but production clips should use the model-supported duration values. |
ratio | 1280:720, 720:1280, 960:960; some workflows also accept 16:9, 9:16, 1:1 | 1280:720 | Use colon-separated WxH ratios when you need predictable Runway frames. Treat 1280:720 as a 1280x720 target, 720:1280 as 720x1280, and 960:960 as 960x960. |
Task flow
Poll the compatibility-format feed route
Authorizations
Bearer token authentication. Use your CometAPI key.
Headers
Optional Runway API version header, for example 2024-11-06.
Body
Webhook URL that receives a POST request with the result when the task completes.
Frame ratio for the generated video. Use colon-separated WxH ratio values such as 1280:720, 720:1280, or 960:960 when you need predictable Runway frames; some workflows also accept aspect-ratio labels such as 16:9, 9:16, and 1:1.
Text prompt describing the video content to generate.
Artistic style for the generated video, e.g. cinematic.
Model version to use, e.g. gen4.
Advanced generation options including duration and camera motion.