Seedance 2.0 Coming Soon
Get notified when it's ready
Generate Video with Seedance 1.5 ProSeedance 2.0: All-round Video Generator
Seedance 2.0 supports four modalities: image, video, audio, and text, offering richer expression options and more controllable generation.

Key Features of Seedance 2.0
Supports multimodal input (Text + Image + Video + Audio)
You can simultaneously upload text prompts, up to 9 images, 3 videos and 3 audio clips, freely combining these materials as "references" or "sources of inspiration." This combination method allows AI to understand explicit references, rather than blindly generating based on a single text.
Reference-driven Generation
The unique Reference (@reference) system allows you to specify the purpose of multiple clips:
- @image1 as character design
- @video1 as camera movement or action rhythm
- @audio1 as music beats or voice-over
This system allows video composition, characters, camera language, rhythm, and more to be "copied and recombined on demand."
Superior image quality and continuity
- Supports multiple resolutions, including 2K cinematic output.
- Generated videos exhibit excellent consistency in character appearance, clothing, lighting style, and scene structure, resolving the "style drift" problem commonly found in traditional AI videos.
- Natural motion synthesis and physically-based movement make characters and shots resemble real video recordings rather than simple collage animation.
Built-in audio and precise lip-sync
Seedance 2.0 can generate matching sound effects, background music, or dialogue, and achieve precise synchronization with the video footage. It supports multiple languages (such as English, Mandarin, Japanese, Korean, Spanish, etc.), which is a key capability for short video content production and music and audiovisual works.
Video Extension and Editing
Not only can Seedance 2.0 generate videos from scratch, but it can also smoothly extend existing short videos, replace characters, modify scene elements, and adjust actions without destroying the remaining content. This feature makes Seedance more like an AI editing tool than a simple generator.
Multi Shot Storytelling
Unlike generating single shots, it can automatically generate coherent story segments containing multiple scenes based on instructions, ensuring consistency in visual style and character design. This means you can generate a short film with a beginning, transitions, and an end using just one prompt.