New

Seedance 2.5 AI Video Generator

Turn a prompt or a single photo into a vertical video of up to 30 seconds, in up to 1080p and with optional sound.

Developer
ByteDance
Length
4–30 seconds
Resolution
720p · 480p · 1080p
Aspect ratios
16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 21:9
Photo to video
Yes
Audio
Optional

What is Seedance 2.5?

Seedance 2.5 is the newest Seedance model from ByteDance and the longest-running one we offer: a single generation can reach 30 seconds. That makes it the pick when a clip needs to tell a small story in one take instead of being stitched together from short shots. It accepts a text prompt, or a prompt plus a first-frame photo, and can generate a soundtrack with the video.

Why use Seedance 2.5

Long single takes

Generate anywhere from 4 to 30 seconds in one go, so a scene can build, turn and resolve without cuts.

Up to 1080p

Pick 480p for drafts, 720p for everyday posts or 1080p when the clip is the final version.

Optional audio

Switch audio on and the video comes with generated sound, ready to post without an extra editing step.

Six aspect ratios

Vertical 9:16 for TikTok, Reels and Shorts, plus 16:9, 1:1, 4:3, 3:4 and 21:9 for everything else.

How to make a video with Seedance 2.5

  1. Open the generatorPick Seedance 2.5 in the model list.
  2. Describe the videoWrite one subject, one setting and one camera move, or add a first-frame photo.
  3. Choose length and ratioKeep 9:16 for TikTok, Reels and Shorts, and set the length and resolution.
  4. Generate and downloadPreview the result, then download or refine it.

Open Seedance 2.5 in the generator →

What to make with Seedance 2.5

Short stories and skits

A 20 to 30 second scene with a beginning, a beat and a payoff, shot as one continuous take.

Photo to video

Upload a portrait, product shot or illustration as the first frame and describe how it should move.

Product and ad teasers

Slow reveals, camera moves and lifestyle shots that need more than the usual 5 to 10 seconds.

Music-led clips

Longer shots with sound on suit dance, travel and mood videos that run with a track.

Seedance 2.5 prompt examples

Cinematic street scene
A woman in a red coat walks through a rainy neon street at night, slow tracking shot from behind, reflections on the wet road, soft ambient city sound.
Use this prompt →
Product reveal
A matte black wireless speaker on a stone pedestal, the camera circles it slowly while soft light sweeps across the surface, minimal studio background.
Use this prompt →
Pet moment
A golden retriever shakes off water in a sunny garden, slow motion, droplets catching the light, handheld camera at dog height.
Use this prompt →

Tips for better Seedance 2.5 videos

  • Describe one subject, one setting and one camera move. Long clips drift when a prompt lists too many ideas.
  • Start at 480p or 720p to check the motion, then re-run the good one at 1080p.
  • For a longer clip, say what changes over time: "starts wide, ends in a close-up".
  • Use a clear, well-lit first-frame photo when you need the subject to look the same throughout.

Seedance 2.5 FAQ

What is Seedance 2.5?

Seedance 2.5 is an AI video model from ByteDance. It creates video from a text prompt or from a prompt plus a starting photo, with clips of up to 30 seconds and optional generated audio.

How long can a Seedance 2.5 video be?

On EveryGen you can choose any length from 4 to 30 seconds in a single generation.

Can I make vertical videos for TikTok and Reels?

Yes. 9:16 is the default ratio in the generator, and the model also supports 16:9, 1:1, 4:3, 3:4 and 21:9.

Does Seedance 2.5 generate sound?

Yes, audio is optional. Turn it on in the generator and the video is created with a soundtrack.

How is Seedance 2.5 different from Seedance 2.0 Mini?

Seedance 2.5 reaches 30 seconds and 1080p. Seedance 2.0 Mini tops out at 15 seconds and 720p, but costs less, so it is the better choice for quick drafts.

How many credits does a video cost?

The price depends on length and resolution and is shown in the generator before you start. Nothing is charged until you press Generate.

More AI video models