字节跳动豆包视频生成模型Seedance 2.0 上线

Core Insights - ByteDance's Doubao App has officially launched the Seedance 2.0 video generation model, which is now available on the app, desktop, and web versions [1][5]. - Users can generate 5-second or 10-second videos by entering relevant prompts in the new "Seedance 2.0" section of the app [1][5]. - The model supports original sound synchronization, multi-shot long narratives, and controllable multi-modal generation, allowing users to create videos with complete native soundtracks [1][5]. Features of Seedance 2.0 - The model can automatically parse narrative logic and generate a sequence of shots that maintain high consistency in characters, lighting, style, and atmosphere [1][5]. - Users can also create their video avatars through a verification process, enhancing creative possibilities [1][5]. - Currently, the model does not support uploading real images as reference subjects [4][8].