Thirty-second brand films
Map a product story from setup to reveal and final brand image with timed action, camera, music, and sound cues.
Coming soon · ByteDance video model
Official preview and Seedance 2.0 workspace
Seedance 2.5 AI Video Generator – Coming Soon
Preview ByteDance's upcoming Seedance 2.5 model, built for longer directed sequences, up to 50 multimodal references, timeline control, and targeted video editing. Seedance 2.5 is not yet available in Open Sora; the generator below starts with Seedance 2.0.
Create now with Seedance 2.0
Seedance 2.5 is coming soon and cannot be selected yet. The complete video workspace below starts with Seedance 2.0 so you can build and test a compatible multimodal workflow today.
Loading the Seedance 2.0 AI video generator…
The embedded generator uses Seedance 2.0, not Seedance 2.5. Generation requires an account and uses credits based on the selected model and settings.
What is Seedance 2.5 and when should you use it?
Understand where Seedance 2.5 fits before spending time on references, prompts, and final-resolution generations.
Seedance 2.5 extends the Seedance workflow from short individual clips toward directed audiovisual sequences. ByteDance's official preview demonstrates longer continuous generation, second-by-second creative instructions, many visual and audio references, multilingual performance, and selective changes to an existing video. Public access, pricing, variants, and final platform controls have not yet been announced for Open Sora.
For filmmakers, advertising teams, ecommerce studios, multilingual marketers, music creators, and narrative video producers planning reference-heavy productions, the practical advantage is not a single headline benchmark. It is the way Seedance 2.5 combines Text, image, video, audio, multimodal reference, timeline direction, and local editing with Text, images, video, audio, scripts, storyboards, product assets, and style guides. That combination determines whether the model can preserve a prepared visual direction or needs to invent most of the scene from language alone.
Use this page to study the announced Seedance 2.5 direction and prepare prompts, references, and timelines. The embedded workspace deliberately defaults to Seedance 2.0 until Seedance 2.5 becomes available in Open Sora.
Seedance 2.5 specifications and supported formats
Use these official capabilities to plan the input, format, duration, and production tier before you generate.
Developer
ByteDance
Model release
Coming soon
Generation modes
Text, image, video, audio, multimodal reference, timeline direction, and local editing
Inputs
Text, images, video, audio, scripts, storyboards, product assets, and style guides
Duration
Up to 30 seconds in one continuous generation
Resolution
Final output options to be confirmed at launch
Aspect ratios
Output formats to be confirmed at launch
Audio
Native audiovisual generation demonstrated in official previews
Best use cases for Seedance 2.5
Start with work where the model's strongest controls create a practical advantage rather than choosing only by maximum resolution.
Reference-led ecommerce ads
Use product packshots, brand rules, motion references, layouts, and music together to guide a commercial sequence.
Multilingual performance videos
Plan dialogue or lyrics, visible performance, camera changes, and audio timing for campaigns across multiple languages.
Selective revisions
Prepare workflows that change a prop, wardrobe detail, background element, or unwanted person without rebuilding every frame.
Official Seedance 2.5 preview video
This 30-second preview comes from ByteDance's official Volcano Ark model page. It demonstrates the announced creative direction but is not a promise of current access in Open Sora.
Official Seedance 2.5 preview
Official source material: Official 30-second Seedance 2.5 preview downloaded from the Volcano Ark model page and optimized for web playback.
Prepare for Seedance 2.5 with Seedance 2.0
Build the brief and reference workflow now, then move it to Seedance 2.5 after the model becomes available.
1
Plan the complete timeline
Write the subject, story beats, camera, sound, and transitions in clear time windows, then assign each reference a specific role.
2
Prototype with Seedance 2.0
Use the generator on this page to test the visual direction, character, product, motion, audio, and reference relationships.
3
Move to 2.5 after launch
When Seedance 2.5 becomes available, switch the model, review its final controls, and refine the prepared brief for longer output.
What Seedance 2.5 is expected to add
Capabilities demonstrated or announced in ByteDance's official preview, presented without implying that the model is already available here.
30-second continuous generation
Direct a complete 30-second sequence in one generation instead of assembling several short clips after the fact.
Up to 50 multimodal references
Combine images, video, audio, scripts, storyboards, products, and style material to define different parts of the brief.
Second-by-second timeline direction
Structure action, camera changes, sound cues, transitions, and the closing image around explicit moments in the timeline.
Targeted video editing
Official previews show local changes that preserve the wider composition, protagonist, timing, and camera movement.
Multilingual audiovisual performance
Coordinate spoken performance, lip movement, music, cuts, and camera direction across multilingual creative concepts.
Longer narrative continuity
Plan connected scenes and transformations while maintaining a readable subject, visual motif, and direction across the sequence.
How to prompt Seedance 2.5 for more controllable video
A strong Seedance 2.5 prompt behaves like a compact production brief: it gives the model a subject, an ordered action, a camera plan, an art direction, and a soundtrack.
Reusable prompt structure
Subject + ordered action + environment + camera + lighting + visual style + timing + dialogue and sound + consistency constraints
Example Seedance 2.5 prompt
“Create a continuous 30-second product film. 0–6s: use Image 1 for the exact product and Image 2 for the desert palette; slow push toward the package as wind moves the sand. 6–15s: follow Video 1's circular camera motion while the product opens and bright citrus particles appear on the beat of Audio 1. 15–24s: transition through three environments while preserving the package geometry and label. 24–30s: settle into a clean hero shot with the exact final tagline, warm backlight, and a short sonic logo.”
Write the timeline first
Divide the 30-second idea into clear time windows and assign one readable action, camera goal, and sound cue to each.
Give every reference one job
Name which asset controls the subject, product, style, motion, composition, voice, rhythm, or transition.
State what editing must preserve
For a local change, explicitly lock the camera, timing, protagonist, lighting, and unaffected areas before describing the revision.
Announced Seedance 2.5 capabilities and pending details
Use confirmed preview information for planning. Final variants, resolutions, aspect ratios, pricing, and API controls remain subject to launch documentation.
| Capability | Seedance 2.5 support |
|---|---|
| Variants | Public variants, pricing, and API details to be announced |
| Generation modes | Text, image, video, audio, multimodal reference, timeline direction, and local editing |
| Inputs | Text, images, video, audio, scripts, storyboards, product assets, and style guides |
| Reference control | Up to 50 multimodal reference assets |
| Duration | Up to 30 seconds in one continuous generation |
| Resolution | Final output options to be confirmed at launch |
| Aspect ratios | Output formats to be confirmed at launch |
| Audio | Native audiovisual generation demonstrated in official previews |
Up to 30 seconds in one continuous generationFinal output options to be confirmed at launchNative audiovisual generation demonstrated in official previews
Plan around the limits of generative video
AI video is most reliable when the prompt gives each shot one readable visual idea. Review important details before publishing and treat the first generation as a directed take that can be refined.
Complex multi-character interaction, fast occlusion, readable text, logos, hands, and exact object counts can still vary between takes. Use clear references, simplify crowded action, and inspect continuity frame by frame.
Higher resolution does not replace art direction. Lock the story beat, composition, movement, and sound at an economical setting first; then move the strongest direction to the premium variant or resolution.
Seedance 2.5 coming-soon FAQ
What is Seedance 2.5?
Seedance 2.5 is ByteDance's upcoming AI video model focused on longer, more directed, reference-rich audiovisual sequences. Official previews highlight 30-second continuous generation, up to 50 multimodal references, timeline direction, multilingual performance, and targeted editing.
Is Seedance 2.5 available in Open Sora now?
No. Seedance 2.5 is listed here as coming soon and is not selectable in the generator. This page will be updated when the model is integrated.
Which model does the generator on this page use?
The embedded generator defaults to Seedance 2.0. You can use it now to prototype prompts, references, motion, audio, and multi-shot direction while waiting for Seedance 2.5.
Can Seedance 2.5 generate a 30-second video?
ByteDance's official preview presents up to 30 seconds in one continuous generation. Exact platform controls and limits will be confirmed when Seedance 2.5 becomes available.
How many references can Seedance 2.5 use?
The announced workflow supports up to 50 multimodal reference assets, including combinations of visual, video, audio, script, storyboard, product, and style material.
Does Seedance 2.5 support local video editing?
Official previews demonstrate targeted changes to selected areas while preserving the broader shot, timing, protagonist, and camera movement. The final editing controls will depend on the released integration.
Does Seedance 2.5 generate audio?
Official previews show coordinated audiovisual output, including speech, music, sound cues, performance, and visible lip movement. Final audio settings will be documented when access opens.
How is Seedance 2.5 different from Seedance 2.0?
Seedance 2.0 is the available production model on this page. Seedance 2.5 is positioned around longer 30-second sequences, many more combined references, finer timeline direction, and selective editing. Keep using 2.0 now and compare the same briefs after 2.5 launches.
Official Seedance 2.5 sources
Model capabilities and media on this page were researched from the developer's official product pages, announcements, and documentation.
Use Seedance 2.0 while Seedance 2.5 is coming
Start building the prompt, reference, motion, and audio workflow today with Seedance 2.0. This page will switch to Seedance 2.5 after the model is integrated.
Try Seedance 2.0 nowBrowse all models