Video generation

Wan 3.0

by Alibaba

The next model expected in Alibaba's Wan line.
Early demos point to 30‑second 1080p clips.

Wan 3.0

Key features

One unbroken push down a rain-slick alley, neon sliding pastStart creating

Expected features, drawn from early demos and hands-on reports rather than published specifications.

Technical specifications

~30s reported

Roughly 30 seconds per generation, from early access.

1080p reported

Described as native in early demos, not upscaled.

No date yet

Alibaba has not announced when it ships.

Not confirmed

Published Wan weights currently stop at Wan 2.2.

Use cases

Longer-form ad spots

A 30-second clip would cover a full broadcast spot in one generation, with no cuts to stitch.

Short drama and skits

Longer single takes suit serialized skits, where a scene needs room past a four-second beat.

Reference-led continuity

If reference-to-video carries over, a character or set stays locked across a whole sequence.

Prompt examples

Long single take

One unbroken push down a rain-slick alley, neon reflections sliding past

Edit prompt

Reference continuity

The same courier crosses three city blocks, morning haze to midday glare

Edit prompt

Product reveal

Studio-lit rotation of a brushed steel kettle on a dark stone slab

Edit prompt

Character performance

A violinist finishes a phrase, lowers the bow, and exhales in low light

Edit prompt

Landscape reveal

Slow aerial pull-back from a river delta as the fog burns off at sunrise

Edit prompt

In-frame typography

Coastal cliff at golden hour with a clean lower-third title card

Edit prompt

Simple pricing

Get started for free today, with the option to upgrade or cancel anytime.

Basic

$9/ month
billed as $0 per year

900 monthly credits

1 user only

All models

Workflows

Standard

$24/ month
billed as $0 per year

3200 monthly credits

1 user only

All models

Workflows

Pro

$45/ month
billed as $0 per year

6200 shared monthly credits

1 user

+ up to 4 more at extra cost

All models

Workflows

Pro Max

$170/ month
billed as $0 per year

24000 shared monthly credits

1 user

+ up to 9 more at extra cost

All models

Workflows

Enterprise

For higher limits

Custom

pricing and billing terms

High-volume credits
Custom seat limits
All models
Workflows
Pricing Gradient

Free

For playing around

$0

forever free

Up to 20 credits
1 user only
Limited models
Workflows
Compare plan details

FAQs

What is Wan 3.0?
Wan 3.0 is the name attached to the next model expected in Alibaba Tongyi Lab's Wan video line. Demo footage circulated in early August 2026 and early testers describe native 1080p output at up to 30 seconds per clip. Alibaba has not published specifications, weights, or a release date, so treat every figure on this page as reported rather than official.
Has Wan 3.0 been released?
No. As of August 2026 there is no official announcement from Alibaba's Wan or Tongyi Lab channels, no repository on the Wan-Video GitHub organization, no weights on the Wan-AI Hugging Face organization, and no wan3.0 entry in Alibaba Cloud Model Studio's video generation documentation. What exists is demo footage and accounts from people given early access.
How long will Wan 3.0 clips be?
Early testers and community posts describe roughly 30 seconds of native single-shot video. For comparison, Alibaba documents the current Wan 2.7 models at 2 to 15 seconds. The 30-second figure has not been confirmed by Alibaba and could change before any launch.
Will Wan 3.0 be open source?
Unknown, and it is the most contested question around the model. Alibaba's published open weights currently stop at Wan 2.2 under Apache 2.0, while the 2.5, 2.6, and 2.7 generations have been commercial API models. Early community reaction to the demos anticipated a closed release, but nothing is confirmed either way.
Will Wan 3.0 generate audio as well as video?
Almost certainly, though it is not confirmed. Recent Wan releases generate audio in the same pass as the picture rather than leaving a silent clip to score afterwards, so a successor dropping that would be a genuine surprise. Expect the same three input modes too: text-to-video, image-to-video, and reference-to-video.
How would Wan 3.0 compare with Seedance 2.5?
Too early to say with any confidence. Both are reported at roughly 30-second native clips, and early testers have run them side by side, but neither model has independent benchmarks and Wan 3.0 has no published specification at all. Any comparison right now rests on a handful of demo clips.

More about Wan 3.0

Wan 3.0 guide: features, prompts, and worked examples

Wan 3.0's expected capabilities, plus how to prompt for long single takes and reference-led continuity, with worked examples.

Other models

Seedance 2.5

ByteDance

ByteDance's next-generation video model. Up to 30s native clips, 50 references, native audio.

Kling 4.0

Kling

Kuaishou's next Kling flagship, expected soon. Longer clips and sharper detail are on the way.

MiniMax H3

MiniMax

MiniMax's multimodal video model. 2K, native stereo audio, clips up to 15s.

Flux 3

Black Forest Labs

Black Forest Labs' world model for image, video, and audio, with sound that matches the action.