Seedance
Seedance 2.0 is ByteDance's latest text-to-video and image-to-video generation model accepting text, images, videos, and audio inputs to produce cinematic multi-shot video with native audio sync and consistent characters.
What is Seedance?
Seedance is a text-to-video (T2V) and image-to-video (I2V) generation model developed by ByteDance's Seed lab. Seedance 2.0 accepts up to 12 assets in a single generation—text, image, video, and audio—producing cinematic-quality videos with native sound including dialogue with lip sync, background music, and ambient effects. Users can generate high-quality videos up to 15 seconds from text, images, video, and audio references with native sound generation, lip sync, director-level camera control, and physics-accurate motion. The model maintains consistent characters, lighting, and style across multiple cuts with multi-shot capabilities.
ByteDance's multimodal AI video generation model with native audio sync
Funding
ByteDance
