Full model
Prioritizes finished output
Use when motion, frame control, and audio belong in the final deliverable.
Seedance 2.0 AI video generation models.
The `seedance-video-generator` Combo uses Seedance 2.0 for text-to-video, image-to-video, first/last-frame control, and synchronized audio. The Fast model accelerates drafts and storyboard clips.
Floatboat · Seedance
Live in a built-in Combo
Seedance
What do you want to finish?
Generate video from this briefAnimate these first and last frames
Run through a built-in Combo Skill
- Live engines
- Seedance 2.0 · 2.0 Fast
- Access
- seedance-video-generator Combo
- Generation modes
- Text · Image · First/last frame
- Audio
- Synchronized audio supported
Seedance model characteristics
A video engine organized around shots, motion, and continuity.
Seedance is not a chat model. Its value appears when a creative brief becomes a controlled moving sequence inside the dedicated Combo.
Fast model
Prioritizes iteration
Use for drafts, shot exploration, and storyboard clips.
Control
Works from frames and briefs
Start from text, an image, or defined first and last frames.
Product surface
Dedicated Combo
The engines do not appear in the main chat dropdown.
Live video engines
Quality and speed variants inside one Combo.
Both models are live through the Combo and do not appear in the main chat dropdown.
doubao-seedance-2-0-260128
Text-to-video, image-to-video, frame control, and synchronized audio.
Live · seedance-video-generator
doubao-seedance-2-0-fast-260128
Fast drafts and storyboard clips.
Live · seedance-video-generator
From brief to moving asset
Build the video around the production decision that matters.
Each workflow begins with a different kind of control rather than the same generic prompt box.
Text-to-video concept
Task input
Describe subject, camera movement, scene, and audio intent.
Processing focus
The full engine turns the production brief into a moving sequence with synchronized audio.
Expected result: A reviewable concept clip.
Animate a key visual
Task input
Provide a source image and define how the scene should move.
Processing focus
Image-to-video preserves the visual starting point while adding motion.
Expected result: A motion asset connected to the original artwork.
Storyboard iteration
Task input
Explore several shot directions before committing to final generation.
Processing focus
The Fast engine produces draft clips for comparing composition and movement.
Expected result: A faster decision on which shot deserves final rendering.
Seedance in Floatboat
Video controls presented as a production workflow.
The Combo provides the correct entry for two live Seedance engines.
Text-to-video
Turns a shot description into a generated moving sequence.
Image-to-video
Uses a source image as the visual basis for motion.
First and last frame control
Defines the visual start and end points of the generated shot.
Synchronized audio
Supports video generation where sound belongs to the resulting sequence.
Choose the engine by the stage of production.
Choose the full Seedance engine when
- The clip is intended as a finished or near-finished deliverable.
- Frame control and synchronized audio matter.
- You already know the shot direction you want.
Choose Fast or another route when
- You are exploring several storyboard directions.
- Speed matters more than final polish in the current round.
- The deliverable is a still image; use the drawing Combo instead.
Video workflow
Brief, generate, then review the moving result.
- 01
Define the shot
Set subject, motion, framing, duration, frames, and audio needs.
- 02
Choose quality or speed
Use the full model for final work or Fast for early drafts.
- 03
Review continuity
Check motion, frames, audio, and brand fit before delivery.
Explore Other Model Families
Compare model families by the kind of work they support. Availability differs by model and entry point.
Seedance · FAQ
What is Seedance 2.0?+
Seedance 2.0 is ByteDance's AI video-generation engine for text-to-video, image-to-video, frame-controlled generation, and synchronized audio.
What does the Fast model change?+
It prioritizes speed for video drafts and storyboard clips, while the full engine is the better choice for finished or near-finished output.
What inputs can Seedance use?+
The live Combo supports text-to-video, image-to-video, and generation controlled by first and last frames.
Can Seedance generate synchronized audio?+
Yes. Synchronized audio is part of the confirmed Seedance 2.0 workflow in the built-in Combo.
What are the Seedance 2.0 model IDs?+
The full model ID is `doubao-seedance-2-0-260128`; the faster variant is `doubao-seedance-2-0-fast-260128`.
Turn a brief and frames into a Seedance video
Use the video Combo for text-to-video or image-to-video, guide opening and ending frames, then choose Fast for draft iterations.