Seedance 2.0

Seedance 2.0 is an AI video model that can generate a multi-shot sequence in one pass. Instead of producing one clip at a time and stitching them together in an editor, you script up to six shots, each with its own prompt and duration, and the model renders them as one continuous video.

Director-level controls

• Up to six shots per video, each with its own prompt and length.
• Up to three reference images to lock character appearance, scene style, and color palette.
• First and last frame control, so you can pin the opening and closing keyframes and let the model fill in the motion between them.
• Element references you can reuse across shots, which keeps a character looking the same from shot to shot.
• Voice sync, so lip movement and expression match an audio track.

Input and output

Inputs can be text, images, video, audio, or any mix. Output is 2K, with configurable aspect ratio and duration for different platforms.

Trying it out

The site offers free credits and does not require a credit card to start. Quality is strongest on short, well-specified shots; long shots with vague prompts tend to drift.