Actions3
- Image Actions
- Video Actions
- Text Actions
Video → Generate Video
AI-generatedSummary
Generate a Seedance video from a text prompt with optional first/last frame images, polling the BytePlus API until completion and returning the video URL.
Inputs
- prompt (required) — Text description of the video to generate
- model — Model to use for video generation; default: seedance-1-0-pro-fast-251015; options: seedance-1-0-pro-fast-251015, seedance-1-0-pro-250528, seedance-1-5-pro-251215, custom
- customModel — Enter a custom model ID (shown when model is set to Custom)
- imageUrl — URL of the first frame image (must be a public https:// URL)
- lastFrameImageUrl — URL of the last frame image (only available for Seedance 1.0 Pro and 1.5 Pro)
- duration — Length of the generated video (2-12s; minimum 4s for Seedance 1.5 Pro; default: 3s; hidden when model is seedance-1-5-pro-251215)
- durationSeedance15 — Length of the generated video (for Seedance 1.5 Pro only; 4-12s; default: 4s; shown when model is seedance-1-5-pro-251215)
- resolution — Output resolution of the generated video; options: 720p, 480p; default: 720p
- aspectRatio — Aspect ratio of the generated video; options: 1:1, 16:9, 21:9, 3:4, 4:3, 9:16, 9:21, adaptive; default: 16:9; adaptive not supported for Seedance 1.0 models or when using reference images
- generateAudio — Whether to generate audio for the video (only available for Seedance 1.5 Pro; default: false)
- additionalOptions — Additional polling and watermark options; includes pollInterval (number, default 2s) and watermark (boolean, default false)
Output shape
object
Polls indefinitely until the video generation task succeeds or fails. Returns the video URL, task ID, elapsed time, and raw API response. Audio is only included for Seedance 1.5 Pro models. Adaptive aspect ratio is rejected if the Seedance 1.0 model or reference images are used.