Seedance 3.0
A cinematic AI video generator combining text, image, video, and audio references
SITE CAPTUREBased on information published on the service's website. Selection and review policy
WHAT IT DOES
What does it do?
Seedance 3.0 generates scenes by assigning different roles to text, images, video, and audio. It can reference up to 50 assets in a scene and carry the characteristics of a person or product consistently across related shots. A single generation can last up to 30 seconds and supports native 4K output.
KEY FEATURES
Key features
- Generate video using text, image, video, and audio references together
- Specify subjects, actions, camera, mood, and sound in a prompt
- Reference up to 30 images, 10 videos, and 10 audio files in 1 scene
- Create up to 30 seconds in a single generation
- Organize people, products, environments, motion, and sound into separate reference groups
- Native 4K output