ByteDance's Seedance 2.0 text-to-video builds the whole frame from a written prompt, with sound in the same pass. Clips run 4 to 15 seconds at 480p, 720p, 1080p or 4K, in seven aspect ratios including adaptive.
Workbench
These examples were written for this endpoint. Replace the prompt and sample media URLs with your own accessible inputs, then check the current quote and model availability before submitting.
- Output5s
neon-bar-tail The clip shown on this page, request and all: five seconds at 720p, 16:9, audio on. Note how the prompt names the film stock, the light and the sound, not just the subject.
- 720p
- 16:9
- audio
- Output5s
forge-of-giants-4k One light source, and it is moving. The brief makes a pour of molten metal the only illumination in frame, so exposure changes across the shot and every body in it is lit by the same travelling source — a much harder problem than a static key. Written from nothing at `1080p`, `16:9`, `duration_seconds` `5`, with `generate_audio` left on so the forge noise is rendered against the picture.
- 1080p
- 16:9
- audio
- Output5s
shore-break-dawn The two clips beside it are neon and interior; this one is flat overcast light, sea fog and a desaturated grey, held on a long lens from a couple of metres away. Framing is written into the brief — hips up, subject filling the frame — because at this distance a wide shot would give you weather and no face.
- 720p
- 16:9
- audio
- Output5s
neon-subway-ride The second clip on this page, request and all. Worth knowing why generate_audio is off: the first attempt named music in the scene and the model's own layer refused to render a soundtrack on music-rights grounds, with nothing billed. Re-running it silent went straight through — if a prompt implies recognisable music, plan to score it yourself.
- 720p
- 16:9
- 5s
This example starts at $0.07 per output second. The Playground updates the estimate when you change the settings.