Wan · Alibaba

Wan 2.6 AI Video Generator

The long-form Wan: 15-second text-to-video with an audio toggle, plus image and video-to-video modes with quick Flash variants.

Text to VideoImage to VideoVideo to Video

Free credits on sign-up · No watermark · Cancel anytime

View pricing →

Wan 2.6 is the endurance runner of Alibaba’s Wan family. Its headline capability is duration — text-to-video stretches to a full 15 seconds, longer than any other Wan mode on Flux 3 AI — and its text-to-video mode ships with a generate-audio toggle, so a single render can arrive with sound already attached.

Five variants are live on the platform: standard text-to-video, image-to-video, and video-to-video, plus Flash editions of the image and video modes that trade a step of fidelity for faster, cheaper renders. Image-to-video takes an upload as your opening frame, video-to-video restyles an existing clip and includes an audio setting for the source track, and every variant supports style presets.

Text-to-video outputs 5, 10, or 15 seconds; the other modes render 5 or 10, all at 720p or 1080p. Image-to-video adds 1:1 square to the usual 16:9 and 9:16. A sensible cadence: rough out motion on a Flash tier, then commit the keeper to the standard variant.

Video previews

See what's possible on Flux 3 AI

Sample generations from the platform — write your own prompt to see what Wan 2.6 does with it.

Cinematic

Environment fly-through sample

Fantasy

Armored rider on a mythical beast in the clouds

Cyberpunk

Glowing eyes reflecting streams of code

Compared

Wan 2.6 vs similar AI video models

Specs from the live catalog — every model below is available on the same account.

SpecWan 2.6Wan 2.7Wan 2.5
Max clip length15s10s10s
Durations5s, 10s, 15s5s, 10s5s, 10s
Resolutions720p · 1080p720p · 1080p720p · 1080p
Aspect ratios333
Audio
Image to video
Reference imagesup to 1up to 9up to 1
Keyframes
Negative prompt

Capabilities reflect what each model supports on Flux 3 AI today.

Why Wan 2.6

What makes Wan 2.6 stand out

Up to 15 seconds

Text-to-video runs 5, 10, or 15 seconds — room for a complete beat rather than a fragment that needs stitching.

Generate-audio toggle

Flip one switch on text-to-video and Wan 2.6 delivers the clip with a generated soundtrack instead of silence.

Video-to-video restyling

Feed in existing footage and a prompt to transform its look, with an audio setting controlling the source sound.

Flash speed tiers

Flash image-to-video and Flash video-to-video return results faster at lower cost — built for iteration rounds.

Square framing on I2V

Image-to-video supports 1:1 output alongside 16:9 and 9:16, covering feed posts without a crop in post.

Popular use cases

Longer narrativesUGC-style adsFootage restylingReels & ShortsPhoto animationMusic clips

Why creators run Wan 2.6 on Flux 3 AI

  • Free sign-up credits cover your first 15-second Wan 2.6 renders before you buy anything.
  • Draft on the Flash tiers, then finish on premium models without leaving one library.
  • Every 1080p download arrives watermark-free, ready to post straight to your feeds.
Specifications

Wan 2.6 specs & capabilities

Output

  • Clip duration5, 10, 15 seconds
  • Resolutions720p · 1080p
  • Aspect ratios16:9 · 9:16 · 1:1
  • Generation modesText to Video · Image to Video · Video to Video

Controls

  • Image to video (start frame)
  • Reference images
  • First & last frame keyframes
  • Audio generation
  • Negative prompt
  • Video input / editing
  • Extend video

Available versions on Flux 3 AI

Every Wan 2.6 endpoint below is selectable in the Create studio.

  • Wan 2.6 (T2V)Wan 2.6 — text-to-video
  • Wan 2.6 (I2V)Wan 2.6 — image-to-video
  • Wan 2.6 (V2V)Wan 2.6 — video-to-video
  • Wan 2.6 Flash (I2V)Wan 2.6 Flash — image-to-video
  • Wan 2.6 Flash (V2V)Wan 2.6 Flash — video-to-video
Prompt ideas

Wan 2.6 prompt examples

Copy one as a starting point, or send it straight to the Create studio.

A street-food vendor assembling tacos through the lunch rush, handheld camera weaving through the queue, sizzling griddle, fifteen seconds of continuous market bustle

Uses the full 15-second duration with the generate-audio toggle on

Animate this portrait: she turns from the window, smiles, and tucks her hair back as the curtains billow, gentle push-in, soft morning light

Image-to-video motion instruction written against an uploaded still

Restyle this skateboarding clip as hand-painted watercolor animation, loose brushstrokes and paper texture, keeping every trick and camera move exactly intact

Tests video-to-video restyling while preserving the source motion

How it works

Create with Wan 2.6 in three steps

1

Describe your idea

Write a prompt or upload a starting image — the more specific the scene, motion, and style, the better Wan 2.6 performs.

2

Pick Wan 2.6 & settings

Choose duration, resolution, and aspect ratio. The studio shows the exact credit cost before you generate.

3

Generate & download

Wan 2.6 renders in the cloud — track progress in your library, then download watermark-free or share with a link.

Open the Create studio
FAQ

Wan 2.6 questions, answered

What is Wan 2.6?

Wan 2.6 is a generation of Alibaba’s Wan video model available on Flux 3 AI in five variants: text-to-video, image-to-video, video-to-video, and Flash editions of the latter two. It’s the Wan release to reach 15-second clips and offers a generate-audio toggle on text-to-video.

How long can Wan 2.6 videos be?

Text-to-video supports 5, 10, or 15 seconds — the longest single generation in the Wan lineup. Image-to-video, video-to-video, and both Flash variants render 5- or 10-second clips.

Does Wan 2.6 create sound?

Text-to-video has a generate-audio toggle: leave it off for silent renders or switch it on to get a soundtrack with the clip. The video-to-video modes include an audio setting for the uploaded clip’s track; image-to-video renders without generated audio.

What are the Wan 2.6 Flash variants?

Flash image-to-video and Flash video-to-video are speed-oriented editions of the standard modes. They keep the same durations, resolutions, and controls but return results quicker at a lower credit cost, which makes them the right tier for drafts and A/B rounds.

Can Wan 2.6 restyle a video I already have?

Yes. Video-to-video mode takes your uploaded footage plus a prompt and re-renders it in the described style, outputting 5 or 10 seconds at 720p or 1080p in 16:9 or 9:16. It also supports a reference image to steer the target look.

What resolutions and aspect ratios are available?

All Wan 2.6 variants render at 720p or 1080p. Text-to-video and video-to-video frame in 16:9 or 9:16, while image-to-video (standard and Flash) additionally offers square 1:1.

Do I need to install anything to use Wan 2.6?

No. Wan 2.6 renders in the cloud on Flux 3 AI — pick the variant in the Create studio, set duration and resolution, and generate from your browser. New accounts get free credits, and finished clips download watermark-free.

Visit the Help Center
Explore more

Related AI models

Browse all models

Start creating with Wan 2.6

Sign up in seconds, get free credits, and put Wan 2.6 to work alongside every other model on Flux 3 AI — one account, one library, no watermarks.

Try Wan 2.6 free