TOOLORA / Local reading edition
ByteDance’s video model family combines text, image, audio and video references for audiovisual generation, extension and targeted editing; current version 2.5.
Seedance is ByteDance Seed’s video-generation family, distinct from Seed language models and the Jimeng application. The current official version is 2.5. Text instructions and image, video or audio references guide subjects, camera movement, motion and sound for joint audiovisual generation, extensions and timestamp-targeted editing. The announcement describes up to 30 seconds per generation with repeated extensions; actual limits and results depend on the service. It is available through Jimeng and Doubao Pro, while the current model directory also links a specific BytePlus playground, used here as the access point. Accounts are required and regions, plans, quotas and model availability vary. The launch article’s forthcoming API statement does not establish a live API.
News summaries are prepared by this site; open the title to read the original.
The announcement introduces 2.5 on Jimeng and Doubao Pro, describing up to 30-second audiovisual generation, repeated extensions, multiple references and timestamp editing. BytePlus ModelArk API access was forthcoming at publication.
ByteDance Seed ↗Product profile →Descriptions summarize public information; consult official websites for features, pricing and regional availability.
