Seedance is ByteDance Seed's family of video generation models, which create short videos from text or images, including stories told over several shots.
In plain words
Seedance 1.0, from June 2025, was trained to handle both text-to-video and image-to-video, and to produce multi-shot videos that keep the same main character and style across cuts. It makes 1080p clips with smooth motion. The newer Seedance 2.5 generates audio and video together and can make clips of up to 30 seconds, with tools for editing and following reference videos.
A quick example
Imagine prompting "a girl finds a lost kitten in the rain, then carries it home; close-up, then wide shot". Seedance would produce one short video with both shots, keeping the same girl and kitten throughout.
Where you'll meet it
- Short narrative and marketing videos.
- Animating still images into video.
Related
Sources
- PaperSeedance 1.0: Exploring the Boundaries of Video Generation Models, ByteDance Seed (arXiv) · read 29 Sept 2026
- DocsSeedance, ByteDance Seed · read 29 Sept 2026
- DocsSeedance 2.5, ByteDance Seed · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.