AI models
Seedance 2.5: 30-second AI video and clip editing
Pick Seedance 2.5 in AI Video Generator for clips of up to 30 seconds guided by images, video and audio, or Seedance 2.5 Edit to change the weather, outfits or style of a clip you already have.
Seedance 2.5 is ByteDance's newer Seedance video model, and in Soundverse it makes 4 to 30 second clips at 480p or 720p from a prompt, a first and last frame, or up to 30 reference images, 10 reference videos and 10 audio clips. Its variant, Seedance 2.5 Edit, re-renders a clip you already have with the change you describe, such as the weather or clothing, and keeps the clip's length and framing. It's built for creators making longer shots, character-led scenes and lip-synced performances.
How it works
Pick Seedance 2.5 or Edit
Open AI Video Generator under Create in AI Tools. In the Model dropdown, choose Seedance 2.5 for a new clip or Seedance 2.5 Edit to change a clip you have.
Describe the scene or change
For Seedance 2.5, describe the scene in the prompt. For Seedance 2.5 Edit, describe only the change in What to change, such as "make it snow" or "change the jacket to red".
Add your media
Add reference images, video and audio from your Library, or turn on Use first / last frame. For an edit, pick an MP4 of 4 to 30 seconds and up to 200 MB from your Library as the Video to edit.
Set the output and generate
Choose 4 to 30 seconds, 480p or 720p, an aspect ratio and Generate audio, then click Generate in Chat; a first frame sets its own framing, and an edit keeps its clip's length and shape. The video lands in your chat and in Videos in your Library.
Why use Seedance 2.5
Clips up to 30 seconds
Seedance 2.5 renders any whole length from 4 to 30 seconds, twice the longest Seedance 2.0 clip, so a full verse or chorus can fit in one shot.
Far more references
Guide Seedance 2.5 with up to 30 reference images, 10 reference videos and 10 audio clips, with each video or audio clip up to 30 seconds.
Lip-sync for a whole verse
Add a portrait as a reference image and a vocal as reference audio, and the character lip-syncs for up to 30 seconds. AI Music Video Generator picks Seedance 2.5 on Auto when the singer lip-syncs.
Edit without reshooting
Seedance 2.5 Edit changes one thing in a clip you already have and saves the result as a new video named after the original, which stays untouched.
Great for
- Longer single shots of up to 30 seconds
- Lip-synced verses and choruses
- Character-replacement trend videos
- Changing the weather or time of day
- Restyling footage you already shot
- Scenes built from many reference images
Frequently asked questions
What is Seedance 2.5?
Seedance 2.5 is ByteDance's newer Seedance video model, available in Soundverse's AI Video Generator together with its editing variant, Seedance 2.5 Edit. In Soundverse, Seedance 2.5 makes 4 to 30 second clips at 480p or 720p from a prompt, a first and last frame, or up to 30 reference images, 10 videos and 10 audio clips.
What is Seedance 2.5 Edit?
Seedance 2.5 Edit is the version of Seedance 2.5 that changes a video you already have instead of making a new one. Pick an MP4 of 4 to 30 seconds and up to 200 MB from your Library, describe the change, such as making it night or turning a jacket red, and Seedance 2.5 Edit re-renders the clip at 480p or 720p with the same length and aspect ratio.
How is Seedance 2.5 different from Seedance 2.0?
Seedance 2.5 makes longer clips, from 4 to 30 seconds against 4 to 15, and takes more references: up to 30 images, 10 videos and 10 audio clips against 9, 3 and 3. Seedance 2.0 renders up to 1080p, while Seedance 2.5 stops at 720p. Seedance 2.0 stays the default in AI Video Generator, so pick Seedance 2.5 when you need its length or references.
How do I use Seedance 2.5 in Soundverse?
Open AI Video Generator, choose Seedance 2.5 or Seedance 2.5 Edit in the Model dropdown, add your prompt and media, and click Generate in Chat. You can also right-click a video lane in AI Studio and choose Generate Video, or ask Agent One by name, such as "make a 20 second clip with Seedance 2.5". Seedance 2.5 is available on the Creator plan and above.
Can Seedance 2.5 make a character lip-sync?
Yes. Add a portrait of the character as a reference image and the vocal as reference audio, up to 30 seconds per clip, and Seedance 2.5 animates the mouth to it; audio without a portrait isn't accepted. A clean vocal from Stem Separator syncs more reliably than a full mix. For a full performance video, AI Music Video Generator picks Seedance 2.5 on Auto when Lip Sync is on. Only use a face and voice you have permission to use.
Does Seedance 2.5 Edit keep my clip's original sound?
No. Seedance 2.5 Edit doesn't keep the source clip's own audio: with Generate audio on, the default, Seedance 2.5 Edit creates a new soundtrack, and with it off, the result is silent. Seedance 2.5 Edit also can't use reference images, a first frame or lip-sync audio, and it can't make a clip longer; to extend a clip of up to 10 seconds, ask Agent One to use Gemini Omni Flash 1.1, which can't edit clips in the EEA, Switzerland or the UK.
Is Seedance 2.5 free?
No. Seedance 2.5 and Seedance 2.5 Edit are available on the Creator plan and above, and each run uses tokens from your plan. The cost depends on the length and resolution: clips guided by reference media cost more per second than prompt or first-frame clips, and an edit is billed on the source clip's length. AI Video Generator shows the estimate before you generate.
Can I use Seedance 2.5 videos commercially?
Seedance 2.5 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 every image, video, vocal and source clip you add. Only lip-sync or show a face or voice you own or have permission to use. Seedance 2.5's content filter can block prompts that involve real people, brand names or song lyrics.
Ready to try Seedance 2.5?
Open AI Video Generator and pick Seedance 2.5 for a new clip, or Seedance 2.5 Edit to change one you already have.