dreamina-seedance-2-0-260128
Dreamina Seedance 2.0 is an audio and video generation model developed by the ByteDance Seed team and integrated into CapCut. This model adopts a dual-branch diffusion Transformer architecture to realize parallel processing of visual and auditory information flows in the generation link, and achieves frame-level precision sound and picture synchronization through the attention/bridging mechanism. The model supports multi-modal input of text, image, audio and video, can parse up to 12 reference files simultaneously, generate videos up to 15 seconds long, and has the ability to synchronously generate native audio and video. Models come with protections like invisible watermarks and C2PA content credentials.
- Context
- 0 tokens
- Max output
- 0 tokens
- Input
- llm video
- Output
- video
Pricing
Input: $7 / 1M tokens
Output: $7 / 1M tokens
480p or 720p : $7 /
480p or 720p : $4.3 /
1080p : $7.700000000000001 /
1080p : $4.7 /
4K : $4 /
4K : $2.4 /
Endpoints
- openai