Wan 3.0

Alibaba video up to 30 seconds: from a description, frames or references — photos, clips and audio. With a soundtrack.

Good for: ads, music clips, story videos, animating photos

Create a videoUp to 10 reference photos5 aspect ratios + adaptiveClips up to 30 sWith sound

Start creating from $0.054per second of video · ~5 ₽

With your video — from $0.054⁠/⁠s (~5 ₽), without video — from $0.054⁠/⁠s (~5 ₽)

For developers →

Create a video

Describe the video

What happens, how the camera moves, the light and style. Refer to uploads: "the woman from photo 1", "music from audio 1".

0/20000

Start from a photooptional

The video starts from this frame. The last frame is optional: the clip smoothly arrives at it.

First frame

Drop a photo here or clickTap to choose a photo JPG, PNG or WebP, up to 30 MB

Last frame optional

Drop a photo here or clickTap to choose a photo JPG, PNG or WebP, up to 30 MB

Add referencesoptional

Photos, clips or audio to build on: a character, a place, motion, music. Clips together up to 15 s. Audio up to 15 s.

Photos — up to 10

Drop a photo here or clickTap to choose a photo JPG, PNG or WebP, up to 30 MB

Clips — up to 5

Drop a video here or clickTap to choose a video MP4 or MOV, 2 s or longer, up to 200 MB

Audio — up to 5

Drop audio here or clickTap to choose audio MP3 or WAV, 2 s or longer, up to 200 MB

Settingsoptional

You can leave them as they are. You pay per second of the clip; the button shows the total.

Resolution ?

Aspect ratio ?

Duration ?

auto (-1)max price first, then recalculated

Video with soundThe AI adds sound to the clip: noises, music, voices. Same price.«AI generated» watermarkAdds a small mark in the corner saying the video is AI-made.Content checkWe check the description and photos before starting. If something is not allowed, no video is made and nothing is charged.

usually about 3 min

How to get a better result

What, where and in what style

"A ginger cat on a windowsill, rain outside, watercolor".

Add light and mood

"Soft sunset", "neon light", "cozy, warm tones".

Not right? Change a word

Refine the description and press "Another version" — every result is new.

About the model

Wan 3.0 is an Alibaba video model. 2 to 30 second clips in 480p, 720p and 1080p: from a description, from the first and last frame, or from references (up to 10 photos, 5 clips and 5 audio tracks). You can turn the soundtrack off — the price stays the same.

Pick «Auto» length and the model decides: the maximum is held and the rest comes back after the video is ready. With your own video, input + output seconds stay within 30.