ByteDance releases Seedance 2.5 video generation model
Summary
ByteDance has launched Seedance 2.5, a video generation model that can produce up to 30 seconds of high-quality video in a single pass. The model supports iterative extension and multimodal reference inputs, including up to 30 images, 10 video clips, and 10 audio clips. ByteDance built it on a unified audio-video architecture to improve long-form storytelling, group scene reconstruction, and editing performance. The company is rolling it out through Jimeng AI and Doubao Professional Edition, and it will also connect API access through Volcano Ark.
AskAI Classifications
Sectors
No sectors detected
Functions
Communication and Collaboration
Collaboration Software
Linked Companies
ByteDance
$1B+