Start with words, a frame or references
Use text-to-video for an open scene, set a first and last frame, or attach up to 9 reference images and 3 reference videos to lock the look.
Why Seedance
These are the controls and limits connected to the current product configuration—not borrowed promises from another provider or model release.
Use text-to-video for an open scene, set a first and last frame, or attach up to 9 reference images and 3 reference videos to lock the look.
Seedance 2.0 renders up to 4K, while 2.0 Fast and 2.0 Mini trade resolution (480p or 720p) for lower cost and quicker drafts.
Pick portrait, landscape, square or classic framing and a length between 4 and 15 seconds on the 2.0 models.
Turn on generated audio for clips that need ambience or dialogue; leave it off for silent b-roll.
How it works
The landing page handles the starting context. Create exposes the full model controls before anything is submitted.
Seedance 2.0 for final quality, 2.0 Fast or 2.0 Mini for low-cost iterations, 1.5 Pro for short 4–12 second clips.
Call out subject movement, camera direction, speed, lighting changes and the mood, and label any references you attach.
Open Create to choose duration, resolution, aspect ratio and audio. The credit cost is shown before you generate.
Best uses
Choose the model around the deliverable, not the novelty of the name. These are the tasks that fit its current inputs and controls.
Turn a clean product still into a short reveal, rotation or atmosphere-led campaign shot.
Create vertical or square clips for feeds, Reels, Shorts and other short-form placements.
Reuse a character, outfit or set across clips by attaching the same reference images.
Test camera movement, scene rhythm and a key moment before a longer production.
Compare models
Use the lighter image model for speed, Pro for resolution and references, and Seedance when the final deliverable needs motion.
| Model | Best for | Input | Output | Action |
|---|---|---|---|---|
| SeedanceThis page | Directed clips with strong reference control | Text, first/last frame, up to 9 images + 3 videos | 4–15s, up to 4K (Seedance 2.0) | View |
FAQ
Short answers based on the model configuration currently available in this workspace.
Seedance is ByteDance's AI video model family. Wan3 AI offers Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini and Seedance 1.5 Pro for text-to-video and image-to-video generation.
Use Seedance 2.0 for the highest quality (up to 4K), Seedance 2.0 Fast or 2.0 Mini for cheaper 480p or 720p drafts, and Seedance 1.5 Pro for short 4, 8 or 12-second clips.
Yes. Upload a JPG, PNG or WEBP as the first frame, add an optional last frame, or attach up to 9 reference images to guide the result.
The Seedance 2.0 models generate 4 to 15 seconds per clip; Seedance 1.5 Pro offers 4, 8 and 12 seconds. Create shows the credit cost after you choose a duration and resolution.
Seedance 2.0 Mini is available on the free plan with your sign-up credits. Seedance 2.0, 2.0 Fast and 1.5 Pro use credits from a paid plan or credit pack.
Wan 3.0 generates longer single takes (up to 30 seconds at up to 1080p) with native audio. Seedance 2.0 is limited to 15 seconds but supports reference videos and up to 4K output. Both are available in the same workspace.
The model opens preselected. Review its live settings and credit cost, add references if needed, and submit when the brief is ready.
Seedance and related product names are trademarks of their respective owners. Availability here refers to the model mapping shown on this page.