Create 4-30-second or smart-duration videos with strict first/last-frame guidance or multimodal image, video, and audio references at up to 1080p.
720p: 39 credits / second × 5s, rounded up for the estimate.
Ready to generate.
Model overview
Create 4-30-second or smart-duration videos with strict first/last-frame guidance or multimodal image, video, and audio references at up to 1080p.
Choose one Seedance 2.5 source workflow before adding media; strict frame guidance and multimodal reference arrays are mutually exclusive.
Creative capabilities
Create 4-30-second or smart-duration videos with strict first/last-frame guidance or multimodal image, video, and audio references at up to 1080p.
Choose an output duration from 4 to 30 seconds, or use smart duration (-1), for flexible scene planning.
Choose 480p, 720p, or 1080p. The workspace initially selects 720p.
Use image, video, audio, and text direction to maintain characters, products, brand assets, visual style, motion, and sound across a sequence.
Use a first frame or first plus last frames as strict endpoints. Strict frame inputs cannot be combined with multimodal reference arrays.
Use up to 30 images, 10 videos, and 10 audio files in multimodal mode, subject to the supported format, size, and aggregate-duration limits.
Model selection guide
Choose one Seedance 2.5 source workflow before adding media; strict frame guidance and multimodal reference arrays are mutually exclusive.
Media source
First frame; optional last frame
Image, video, and audio arrays
Mutual exclusion
No reference arrays
No strict first/last frames
Resolution
480p/720p/1080p
480p/720p/1080p
Duration
4-30 seconds or -1
4-30 seconds or -1
| Capability | Strict-frame workflow | Multimodal-reference workflow |
|---|---|---|
| Media source | First frame; optional last frame | Image, video, and audio arrays |
| Mutual exclusion | No reference arrays | No strict first/last frames |
| Resolution | 480p/720p/1080p | 480p/720p/1080p |
| Duration | 4-30 seconds or -1 | 4-30 seconds or -1 |
Production workflow
Move through shot direction, reference preparation, output settings, and result review in a clear sequence.
Open the video workspace and choose the Seedance workflow.
Write the subject, action, environment, camera, light, atmosphere, and pacing.
Anchor identity, product details, composition, wardrobe, and scene style with useful images.
Review motion, visual consistency, framing, rhythm, and storytelling before refining the next version.
Organize either strict frame guidance or multimodal references so the model can follow characters, products, camera language, sound, and scene structure more clearly.
01
Use images, videos, audio, and written direction to clarify visual identity, framing, motion, rhythm, and cinematic language.
02
Anchor the start with one first frame and optionally the end with one last frame; a last frame requires a first frame.
03
Choose strict frames or multimodal references for each task. The current Seedance 2.5 generator does not expose a separate video-edit workflow.
WAN30 turns model selection, prompts, references, output settings, credit estimates, results, and iteration into one creator-facing workflow.
01
Start from this model page without rebuilding the creative brief in a separate tool.
02
Move between Seedance 2.5, Wan 3.0, Video Prime, and Wan 2.7 without leaving the WAN30 workspace.
03
Validate prompts, references, duration, framing, audio direction, and output settings before repeating a workflow.
04
Review the live cost estimate and current account information before submitting each generation task.
05
Manage supported inputs, generation settings, examples, results, and history in a visual interface.
06
Use this page to compare features, understand current limits, review FAQ answers, and return directly to the workspace.
Flexible duration, up to 1080p output, and richer references support both brand content and cinematic storytelling.
01
Use 4-30 seconds or smart duration to combine product references, campaign direction, feature moments, and a clear ending.
02
Develop product showcases, lifestyle scenes, seasonal campaigns, and marketplace assets at the highest supported output tier.
03
Keep products, characters, colors, wardrobe, locations, and brand styling connected across changing shots.
04
Turn scripts, character references, mood boards, camera ideas, and sound direction into a more complete cinematic sequence.
05
Use recurring character and brand references for fashion, beauty, gaming, lifestyle, and social-media concepts.
06
Plan content around licensed characters and approved brand assets while keeping usage rights and publishing review outside the generation step.
Start a new project
Return to the workspace to shape a prompt, prepare references, compare a direction, and start the next generation task.