ByteDance Releases Seedance 2.5, Supporting 30-Second Videos with 50 Materials

ByteDance's Seed team has released the video model Seedance 2.5, increasing the single generation duration from 15 seconds to 30 seconds, with the ability to further extend the video. This model can arrange multiple shots and a complete storyline within 30 seconds, allowing users to repeatedly continue writing based on existing results, maintaining consistency in characters, scenes, sounds, and narrative rhythm, ultimately stitching together several minutes of video. Seedance 2.5 can input up to 30 images, 10 video clips, and 10 audio segments at once. The new model supports timestamp control and modification of the video, enabling users to specify actions appearing in certain seconds, switch shots, and adjust characters, sounds, or camera movements without needing to redo entire segments. Seedance 2.5 is gradually being launched on Jiemeng AI and Doubao Pro, with the API set to integrate with Volcano Ark soon.