Seedance 2.0

Seedance 2.0 is ByteDance’s unified audio-video generation model, accepting text, images, audio, and video as prompts and references. It is designed for multi-shot storytelling, consistent subjects, complex interactions, stable motion, realistic physics, and reference-guided editing while generating sound and visuals together. Creators access it through ByteDance products such as Dreamina and through the BytePlus ModelArk API, with limited free credits available in the consumer interface.