Skip to content

Seedance 2.5 · 2.0 · 2 Fast · 2 Mini · 1.5 Pro · up to 4K

Seedance 2.5 & 2.0 — ByteDance's AI video model

Make multi-shot videos with sound from text, an image or up to 30 references — up to 30 seconds and 4K on Seedance 2.5, ByteDance's newest video model.

Try
Courier on a bicycle in a lantern-lit market with jacket and hair moving
  • Up to 30 s (2.5)
  • Up to 4K
  • Sound in the same pass
  • 30 reference images

Updated by the

Made with Kenerate

Quick answer

Updated

What is Seedance 2.5?

Seedance 2.5 is ByteDance Seed's AI video model, released in July 2026, that generates audio and video together for up to 30 seconds per clip from text, images or up to 30 reference images. In Kenerate AI's studio you can run seven Seedance versions, with edit, extend and up to 4K on the 2.x models.

Key facts

Longest clip
30 s · 2.5
Top resolution
4K
Reference images
Up to 30
Sound
Yes
Versions here
2.5, 2.0, 2 Fast, 2 Mini, 2 Turbo, 1.5 Pro, 1.0

Real demo

Seedance 2.5 vs Wan 3 — same image, same prompt

Switch between the two and turn the sound on. Same start image and motion prompt, two models.

  • Seedance 2.5. Seedance 2.5 image to video, 5 s at 720p with sound.
  • Wan 3. Wan 3 image to video, 5 s at 720p with sound.

Both are unedited outputs from Kenerate's video studio.

Try Seedance 2.5

Real results

Real Seedance 2.5 clips

Five Seedance 2.5 image-to-video clips (5 s, 720p, with sound), each from a Seedream 5.0 Pro still.

Real Seedance 2.5 clips (5 s, 720p, with sound). Open one to watch it and copy both prompts.

How to

How to use Seedance on Kenerate

Time needed
Steps
4 steps
  1. Step 1

    Open the video studio

    Pick Seedance 2.5 — or 2.0, 2 Fast, 2 Mini, 2 Turbo or 1.5 Pro.

  2. Step 2

    Choose how to start

    Text, a start image (plus an end frame), or up to 30 reference images, 3 videos and 10 audio clips on 2.5.

  3. Step 3

    Write the prompt

    Describe action, camera and sound. With references, name them in the prompt.

  4. Step 4

    Set length and resolution

    4–30 s on 2.5, 4–15 s on 2.0; 480p to 4K.

Kenerate AI Video Studio in Text to Video mode with a lighthouse scene prompt typed in

Open Video Studio in Text to Video mode and write the shot — subject, action, camera move and light.

1 / 5 · Describe the scene

Watch the tour · 1:19

Text, image & reference to video, models, edit, extend & chat

Chapters

Engines

Seedance versions in the studio

From the studio's own model list.

  • Seedance 2.5 title card — a multi-shot cinematic triptych of a heist sequence in crisp 4K
    Newest

    ByteDance

    Seedance 2.5

    Input
    Text, image or 30 images + 3 videos + 10 audio
    Output
    4–30 s · 480p–4K · sound

    Best for

    Longest multi-shot clips; edit and extend

    Open Seedance 2.5
  • Seedance 2.0 title card — a cinematic film still of a lone astronaut walking through a field of sunflowers

    ByteDance

    Seedance 2.0

    Input
    Text, image or 9 images + 3 videos + 3 audio
    Output
    4–15 s · 480p–4K · sound

    Best for

    Seedance 2 quality

    Open Seedance 2.0
  • Seedance 2 Fast title card — a fast-cut cinematic chase still through neon streets

    ByteDance

    Seedance 2 Fast

    Input
    Same as 2.0
    Output
    4–15 s · 480p–4K · sound

    Best for

    Faster Seedance 2

    Open Seedance 2 Fast
  • Seedance 2.0 Mini title card — a compact cinematic film still of a lighthouse at dusk

    ByteDance

    Seedance 2.0 Mini

    Input
    Same as 2.0
    Output
    4–15 s · 480p–4K · sound

    Best for

    The compact Seedance 2 tier

    Open Seedance 2.0 Mini
  • Seedance 2 Turbo title card — reference images quickly turning into a film still

    ByteDance

    Seedance 2 Turbo

    Input
    Text or references
    Output
    4–15 s · 720p or 1080p · sound

    Best for

    Quick reference clips

    Open Seedance 2 Turbo
  • Seedance 1.5 Pro title card — a film still of waves crashing on cliffs with visible sound waves

    ByteDance

    Seedance 1.5 Pro

    Input
    Text or image (+ end frame)
    Output
    4–12 s · up to 1080p · sound

    Best for

    The first Seedance with audio

    Open Seedance 1.5 Pro
All models (7)

The clips above were made with Seedance 2.5; the comparison uses Wan 3.

Tips

Getting the most from Seedance

From ByteDance's guides and our tests.

  • Describe sound too — Seedance generates it with the picture.

  • Write shots in order ('Shot 1… Shot 2…') for multi-shot clips.

  • Name each reference in the prompt so the model knows its role.

  • Test on 2 Fast or 2 Mini, finish on 2.5.

  • Use an end frame to land the final pose.

  • Real people as references need their permission.

Example uses

How creators use Seedance 2.5 & 2.0

Illustrative examples of typical workflows, not customer reviews.

  • Hands pouring rosetta latte art in a sunlit café

    Café owner

    A café owner could turn a latte-art photo into a clip with milk-steamer hiss and café chatter for a menu reel.

  • Skateboarder mid-ollie over stairs in a sunny concrete plaza

    Action-sports filmmaker

    An action-sports filmmaker could write the shots in order and use an end frame to land a skate trick exactly.

  • Woman in a long coat on a rainy rooftop overlooking a city at night

    Short-film director

    A short-film director could feed reference images of one character to keep them consistent across a 30-second multi-shot scene.

FAQ

Seedance 2.5 & 2.0 questions

What is Seedance?

ByteDance Seed's AI video model. Seedance 2.0 and 2.5 generate audio and video together from text, images, video and audio references.

Link to this answer

What's new in Seedance 2.5?

Up to 30 seconds per generation (2.0 did 15), up to 30 reference images, 10 videos and 10 audio clips per ByteDance, plus timestamp-level editing. It was released on 31 July 2026.

Link to this answer

How do I use Seedance 2.0?

Upload your references, name each one in the prompt (like '@Image 1' in ByteDance's own apps), describe the shots and sound, then pick duration, ratio and resolution.

Link to this answer

How do I write a Seedance 2.0 prompt?

Describe the action, camera and sound; write multi-shot clips in order; say what each reference is for.

Link to this answer

Can Seedance make 4K video?

In Kenerate's studio, Seedance 2.5, 2.0, 2 Fast and 2 Mini offer 4K.

Link to this answer

Does Seedance make sound?

Yes — Seedance 1.5 Pro and every 2.x model generate audio with the video.

Link to this answer

Seedance 2.5 vs MiniMax H3: which should I use?

Seedance 2.5 for length and resolution: up to 30 seconds and 4K, with up to 30 reference images. MiniMax H3 Pro tops out at 15 seconds and 2K, and takes 9 images, 3 videos and 3 audio clips as references. Both work from text, an image or references in the same studio, so you can run one prompt on each and compare.

Link to this answer

Is Seedance 2.5 free?

In Kenerate's studio, Seedance runs on one-time credit packs that never expire. ByteDance's own apps (Dreamina, CapCut) have their own plans.

Link to this answer

In depth

Seedance 2.5 & 2.0, explained

Seedance, version by version

1.0 (June 2025) made 1080p clips with 2–3 cuts in 10 seconds. 1.5 Pro (December 2025) was the first to generate sound with lip-sync. 2.0 (February 2026) took text, image, video and audio references for 15-second multi-shot clips. 2.5 (July 2026) reached 30 seconds and 30 reference images.

Explore

Every Kenerate video tool

Generators, tools, models and guides — all in one studio.

See more of Kenerate AI in your Google resultsAdd kenerateai.com as a preferred source — Google shows more from your preferred sources in Top Stories and AI answers.

Try Seedance 2.5 on Kenerate

One studio for images and video.

Tried it? Tell us how it went