AI video model · by ByteDance
Seedance AI video generator
Seedance is ByteDance's video model family. MotionLab offers the Seedance tiers that are enabled below, from lighter draft tiers to the full models, including reference-to-video where you guide a shot with images of your subject.
Seedance variants available now
These are the Seedance variants currently enabled on MotionLab and the settings each one accepts. The table is generated from the live model catalog, so it changes when a variant is added, updated or paused.
| Variant | Modes | Duration | Resolution | Aspect ratios | Audio / end frame |
|---|---|---|---|---|---|
| Seedance 2 Mini | Image to video · Reference to video · Text to video | 4–15s | 480p · 720p | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 21:9 | Native audio · End frame |
| Seedance 2.5 | Reference to video · Image to video · Text to video | 4–30s | 480p · 720p · 1080p | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 21:9 | Native audio · End frame |
| Seedance 2.0 Standard | Reference to video · Image to video · Text to video | 4–15s | 480p · 720p · 1080p | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 21:9 | Native audio · End frame |
| Seedance 2.0 Fast | Reference to video · Image to video · Text to video | 4–15s | 480p · 720p | 1:1 · 16:9 · 9:16 · 4:3 · 3:4 · 21:9 | Native audio · End frame |
What to use Seedance for
- Reference-to-video: keep a character, product or outfit consistent by giving reference images.
- Fast, low-cost drafts on the lighter tiers while you search for the idea.
- Multi-shot social content where several clips need to feel like one campaign.
- Image-to-video for stylised and illustrated looks.
How to use Seedance on MotionLab
- Sign in and open the video creation page.
- Choose a Seedance variant in the model picker. Only the variants listed above are shown.
- Write your prompt, add a start image or references if the mode needs them, and pick duration, resolution and aspect ratio.
- Check the credit cost shown for your settings, then generate. Failed jobs are refunded automatically.
- Download the result or keep refining it in your library.
Seedance prompt tips
- In reference-to-video, say which reference is the subject and what it should do in the scene.
- Draft on a lighter tier with the exact prompt you plan to finish with, so results transfer.
- Keep references clean: one subject per image, neutral background, no text overlays.
- Describe camera and action separately — 'the camera orbits slowly; she turns toward it'.
More technique in the MotionLab prompt guide and the camera motion guide.
Compare Seedance with other models
Seedance questions
What is the difference between the Seedance tiers?
The lighter tiers trade some fidelity for lower cost and are good for drafts; the full tiers are for final takes. The specs table lists every tier that is enabled right now with its supported settings.
What is reference-to-video?
Instead of a single start frame, you give the model images of the subject you want to keep consistent, and it builds a new shot around them.
How much does Seedance cost on MotionLab?
Seedance generations spend MotionLab credits from the same wallet used by every model on the site. The credit cost for your exact settings is shown before you generate, and a job that fails is refunded automatically. Credit packs are listed on the pricing page.
Do I need a separate Seedance account or API key?
No. MotionLab runs Seedance for you: sign in, pick the model in the create page and generate. Your results are saved to your MotionLab library next to work from every other model.