Video Gen Editor

Wan 3.0: Alibaba's All-in-One AI Video Model

Wan 3.0 is Alibaba's all-in-one video model. It accepts more kinds of input than anything else in the studio: text, images, video, audio, a document or even a web page. It's built for longer, reference-heavy work where characters and products have to stay consistent, and it generates sound along with the picture.

Free sign-up · see the cost before you generate

Wan 3.0 specs

Made by
Alibaba
Clip length
2 to 30 seconds
Resolutions
480p, 720p, 1080p
Shapes
16:9, 9:16, 1:1, 4:3, 3:4
Starts from
Text, Image, Video, Audio
Extra inputs
First and last frames, up to 10 reference images, up to 5 reference videos, up to 5 reference audio clips, a document or web page
Sound
Generated with the video

Wan 3.0 pricing

Wan3.0 price per resolution
ResolutionPer second5-second clip10-second clip
480p$0.05$0.25$0.50
720p$0.10$0.50$1.00
1080p$0.20$1.00$2.00

These are the published per-second rates. The composer shows the exact estimate for your settings before you generate, and failed generations aren't charged. Prices last checked September 24, 2026. Credits come with our plans.

What it's good at

  • Clips of 2 to 30 seconds, or let the model choose the length.
  • Up to 10 reference images, 5 reference videos and 5 audio clips in one generation.
  • Reads a document or a web page as context, which is handy for turning an article or a product page into a video.
  • First and last frame control for image-to-video.
  • Native audio, plus an optional prompt extender that fleshes out short prompts.
  • Video-to-video, editing and extending as well as generating from scratch.

Best for

  • Explainers and product demos

    Point it at the page or the doc and describe the video you want.

  • Music-led sequences

    Reference audio sets the mood. See the AI music video generator.

  • Recurring characters

    Reference images keep faces and outfits steady across a series of shots.

  • Scripts

    Long, multi-part prompts in text to video AI.

Things to know

  • It tops out at 1080p.
  • With lots of references, prompts get complicated. Start with one or two and add more only if you need them.
  • Frame images can't be mixed with reference images, videos or audio in the same generation.
  • Wan 3.0 Prime gives the same results faster, for more money ($0.14 vs $0.10 a second at 720p).

A Wan 3.0 prompt to try

Turn the attached product page into a 10-second explainer: the bottle turns slowly on a marble counter while three short benefit lines appear on screen one after another. Clean studio light, soft upbeat music.

Attach the page as a URL input with the + button. Wan uses it as context, so you don't have to paste the copy into the prompt.

Alternatives to Wan 3.0

Compare every AI video model

Use Wan 3.0 for

Frequently asked questions

What's the difference between Wan 3.0 and Wan 3.0 Prime?

They're the same model with the same inputs and the same output quality. Prime runs on faster inference, so you wait less, and it costs more: from $0.068 a second against $0.05 for Wan 3.0.

Can Wan 3.0 make videos with sound?

Yes. Switch audio on and describe the sounds you want, from ambience to music, in the prompt.

Can Wan 3.0 use a document or a website?

Yes, one document or one URL per generation. It reads it as context for the video, which suits explainers and product videos.

How much does Wan 3.0 cost?

From $0.05 a second at 480p to $0.20 at 1080p. A 5-second 480p clip costs about $0.25.