ByteDance Seedance 2.5 generates 4-30 second videos up to 4K from a text prompt, with native audio and optional reference images, videos and audio.

Generate native audio synchronized with the video

Scene, action, camera movement, lighting and mood. Refer to references as @Image 1, @Video 1, @Audio 1 (upload order)

Up to 10 audio URLs guiding soundtrack, ambience or audio style (30s combined max)

Up to 30 image URLs guiding style, characters or composition

Up to 10 video URLs guiding motion, pacing and camera (30s combined max). Billed at a lower per-second rate on output plus reference duration

Output video resolution: 480p ($0.18/s), 720p ($0.36/s), 1080p ($0.90/s), 4k ($1.80/s)

Seedance 2.5 Text-to-Video