Seedance 2.5 was announced by ByteDance on June 23, 2026, with enterprise preview access and broader availability expected in early July. Follow the release while creating videos today with our currently supported Seedance 2.0 workflow.
Seedance 2.5 is not yet available in this generator. Current generation uses Seedance 2.0.
No Videos Generated
Seedance 2.5 is ByteDance's newly announced AI video model, focused on richer multimodal control, longer-form generation, and more production-ready visual storytelling. The model is not yet connected to this site; Seedance 2.0 remains the active generation option.
The announced model targets longer clips, including up to 30-second generation workflows.
Use text, images, audio, and video references to guide style, characters, camera language, and scene continuity.
Direct camera movement, lighting, pacing, and visual narrative using filmmaking language.
Maintain characters, objects, environments, and style across complex shots and transitions.
Prepare reusable video workflows now with Seedance 2.0 and adapt them when Seedance 2.5 becomes publicly available.
Explore the next generation of multimodal video while using the available Seedance 2.0 models for practical generation today.
Describe a scene, action, camera movement, and visual style to generate a video from text.
Animate a reference frame while preserving its subject, composition, and visual identity.
Use an existing clip as a motion and structure reference for a new generated result.
Create landscape, portrait, square, and additional layouts for film, social, and advertising workflows.
Choose supported output resolutions based on the selected Seedance 2.0 model and generation mode.
Track generation progress and revisit completed AI video tasks from your account.
Examples of the video concepts and production workflows that creators can explore with the Seedance model family.

Develop cinematic environments, visual effects concepts, and narrative shots from detailed prompts.

Create surreal motion, atmospheric sequences, and visual concepts for music content.

Visualize historical, scientific, and abstract subjects when suitable footage is unavailable.

Prototype campaign shots, product motion, and stylized environments without a full physical production.
Release, early access, capability, and current generator information.
Follow release updates and start building your cinematic AI video workflow today with the currently supported Seedance 2.0 generator.