字节跳动发布Seedance 2.0，豆包、即梦接入

Core Viewpoint - ByteDance has launched its latest video generation model, Seedance 2.0, which enhances user experience in creating AI-generated videos through its products Doubao and Jimeng [1] Group 1: Product Features - Seedance 2.0 supports four modalities of input: image, video, audio, and text, allowing for a richer expression and more controlled generation of content [1] - Users can create digital avatars by completing a real-person verification process through recording, enabling them to generate AI videos featuring their likeness [1] - The model allows for a more natural and efficient creative process, resembling that of a true "director," as it expands beyond text-based prompts [1] Group 2: User Experience - Doubao App and Jimeng App support real-person appearances, while the desktop and web versions do not currently allow for the upload of real human face materials [1] - The model has gained global attention due to its multi-modal references and precise controllability, following a limited internal testing phase [1]