AI video generator

Generation settings

Model

Settings for Seedance 2.0 · from image

Prompt required0 / 5000Text prompt describing the desired video motion. Optional but recommended.

Image requiredFirst-frame image URL, Base64, or asset reference (asset://<ASSET_ID>). The video starts from this image. Formats: jpeg/png/webp/bmp/tiff/gif, dimensions (300,6000)px, aspect ratio

Last Image optionalLast-frame image URL, Base64, or asset reference. The video transitions from the first frame to this last frame. Same format requirements as image.

DurationVideo duration in seconds (4-15), or -1 for model to choose automatically.

ResolutionVideo resolution. "4k" is native UHD (3840×2160, 16:9) available on the full Seedance 2.0 models only (not Fast/Mini); 4K output is 10-bit, H.265/HEVC encoded and may not play in a

RatioAspect ratio. 'adaptive' matches the first frame image aspect ratio.

Bitrate Mode

Output video bitrate mode. 'high' encodes at a higher bitrate for a crisper, larger file; 'standard' uses the normal bitrate. Does not affect token cost.

Generate Audio

Whether to generate synchronized audio.

Return Last Frame

Whether to return the last frame as a separate image.

Cost when signed in~45 credits · ~150s

A simulated preview, kept in this browser only. Sign in to generate for real with credits.

Nothing generated with Seedance 2.0 · from image yet

Add a start frame and describe only the motion. Keep the camera move to one verb.