Seedance 2.0
AI video generation model supporting multimodal references and natural-language control
About Seedance 2.0
Seedance 2.0 is a multimodal AI video generation model that combines images, videos, audio, and text to create videos, with natural-language descriptions controlling motion, effects, camera, characters, scenes, and sound. The official FAQ describes support for multiple input materials, 4–15s videos, multiple aspect ratios, and up to 1080p output, making it suitable for content creation that needs reference materials and controllable generation.
Best for
Best use cases
Create videos by combining images, videos, audio, and text
Generate videos from reference materials
Control camera, motion, and sound with natural language
Compare before choosing
The official comparison page states that Seedance 2.0 can use natural-language descriptions to reference motion, effects, camera movement, characters, scenes, and sound.
Features
Multimodal input
Supports image, video, audio, and text inputs.
Natural-language control
Natural language can describe motion, effects, camera movement, characters, scenes, and sound.
Flexible video output
Supports 4–15s videos, multiple aspect ratios, and up to 1080p.
Reference-based creation
Videos can be used as reference materials.
Pros
- Supports multimodal image, video, audio, and text inputs
- Natural language can control motion, effects, camera, and more
- Supports multiple aspect ratios and output up to 1080p
Cons
- Input materials have quantity and duration limits: up to 9 images, 3 videos, and 3 audio files; video and audio totals must each be no more than 15 seconds.
- Generation of NSFW, explicit, or other adult content is not supported; related prompts and uploaded content are blocked by the safety filter.
Pricing Plans
Basic
- Access to core features
Basic
- Access to core features
- 9,600 credits/year
* Pricing information is for reference only. Please refer to the latest announcement on Seedance 2.0's official website.