AI models
Seedance 2.0: the default AI video model in Soundverse
Seedance 2.0 runs AI Video Generator unless you pick another model. Make 4 to 15 second clips at up to 1080p in six aspect ratios, guided by images, video or a vocal your character lip-syncs to.
Seedance 2.0 is ByteDance's video model and the default in Soundverse's AI Video Generator: it turns a prompt, a first and last frame, or reference images, video and audio into a 4 to 15 second clip at 480p, 720p or 1080p. It also powers Lip Sync and Multi-Scene Video Builder, and Agent One uses it when you ask for a new clip of up to 15 seconds without naming a model. It's built for creators who need flexible shapes, 1080p clips and lip-synced performance shots.
How it works
Open AI Video Generator
Open AI Video Generator under Create in AI Tools; Seedance 2.0 is already selected in the Model dropdown. In AI Studio, right-click a video lane and choose Generate Video.
Describe the shot
Write the subject, action, camera and look of the clip. Enhance can polish the prompt once you've written it.
Add references or frames
With Use media as reference on, add up to nine reference images, up to three reference clips totaling 15 seconds and up to three vocal clips from your Library. Or turn on Use first / last frame to start and end on images you choose.
Set the output and generate
Choose 4 to 15 seconds, 480p, 720p or 1080p, an aspect ratio and whether to generate audio, then click Generate in Chat. The clip arrives in your chat and in Videos in your Library.
Why use Seedance 2.0
Six aspect ratios
Seedance 2.0 renders 16:9, 9:16, 1:1, 4:3, 3:4 and 21:9, so one model covers widescreen, vertical, square and ultra-wide shots.
1080p for longer clips
Seedance 2.0 is the only model in AI Video Generator that renders 1080p clips longer than 10 seconds, at any length up to 15 seconds.
Lip-sync from a vocal
Add a portrait as a reference image and a vocal of up to 15 seconds as reference audio, and Seedance 2.0 makes the character lip-sync to it.
The default across tools
Seedance 2.0 runs Lip Sync and every scene in Multi-Scene Video Builder, and Agent One uses it for a new clip of up to 15 seconds when you don't name a model.
Great for
- Vertical clips for Reels and Shorts
- Square and 4:3 clips for social feeds
- Lip-synced performance shots
- 1080p b-roll for a release
- Animating artwork from a first frame
- Matching the motion of a reference clip
Frequently asked questions
What is Seedance 2.0?
Seedance 2.0 is ByteDance's video model and the default model in Soundverse's AI Video Generator. In Soundverse, Seedance 2.0 turns a prompt, a first and last frame, or reference images, video and audio into a 4 to 15 second clip at 480p, 720p or 1080p, in six aspect ratios.
How do I use Seedance 2.0 in Soundverse?
Open AI Video Generator, where Seedance 2.0 is already the default in the Model dropdown, describe the shot, set the output and click Generate in Chat. Agent One also uses Seedance 2.0 when you ask for a new clip of up to 15 seconds without naming a model, and Lip Sync and Multi-Scene Video Builder run on Seedance 2.0 too. Seedance 2.0 is available on the Creator plan and above.
How long can a Seedance 2.0 video be?
Each Seedance 2.0 run makes one clip of 4 to 15 seconds, at 480p, 720p or 1080p. For one clip of up to 30 seconds, pick Seedance 2.5, which renders up to 720p. For a longer piece, Multi-Scene Video Builder films up to 10 Seedance 2.0 scenes of 4 to 15 seconds each and joins them into one video.
Can Seedance 2.0 make a character lip-sync?
Yes. In AI Video Generator, add a portrait of the character as a reference image and the vocal as reference audio, up to 15 seconds per clip, and Seedance 2.0 animates the mouth to it. Lip Sync offers the same thing in a simpler panel. A clean vocal from Stem Separator syncs more reliably than a full mix. Only use a face and voice you have permission to use.
Why won't Seedance 2.0 animate my photo?
Seedance 2.0's safety filter blocks realistic human faces, so animating a real photo of a person often fails, and rewording the prompt won't help. Use an illustrated or stylized character, choose shots that don't show the face, or pick Veo 3.1 or Gemini Omni Flash 1.1 for a photoreal character. Real people, brand names and song lyrics in a prompt can also trigger the filter.
How is Seedance 2.0 different from Seedance 2.5?
Seedance 2.5 makes longer clips, up to 30 seconds, and takes far more references: up to 30 images, 10 videos and 10 audio clips. Seedance 2.0 stops at 15 seconds and up to 9 images, 3 videos and 3 audio clips, but Seedance 2.0 renders 1080p, while Seedance 2.5 tops out at 720p. Seedance 2.0 stays the default in AI Video Generator.
Is Seedance 2.0 free?
No. Seedance 2.0 is available on the Creator plan and above, and each run uses tokens from your plan. The cost depends on the clip length and resolution, and adding a reference video raises it; AI Video Generator shows the estimate before you generate. You can sign up for Soundverse free and explore the core tools before you upgrade.
Can I use Seedance 2.0 videos commercially?
Seedance 2.0 runs on paid Soundverse plans, and paid plans add a commercial license; the Pricing page lists what each plan covers. You need the rights to anything you add as input, including reference images, video and audio and first and last frames. Only use a face or voice you own or have permission to use.
Ready to make a clip with Seedance 2.0?
Open AI Video Generator, where Seedance 2.0 is already selected, and describe your first shot.