Project MonetRequest demo
Home/Blog/Wan 3.0: API, Pricing, 30-Second Video, Native Audio & How It Works

AI · Project Monet Briefing

Alibaba Wan 3.0: API, Pricing, 30-Second Video and Native Audio

Alibaba's publicly documented preview API combines up to 30-second video, native audio, editing, extension and unusually broad reference inputs behind one model.

Published 2026-08-28 · Updated 2026-08-28 · By Project Monet Editorial Team

Project Monet editorial graphic for Alibaba Wan 3.0 with 30-second video, native audio and multimodal references

01

What is Wan 3.0?

Alibaba Cloud introduced Wan 3.0 in August 2026 as an all-in-one generative-video model in Model Studio. The public model ID is wan3.0-video, with a Prime variant listed separately.

One model covers text-to-video, image-to-video, first/last-frame generation, multimodal reference generation, video editing and video extension.

02

Is Wan 3.0 available?

Yes, through Alibaba Cloud Model Studio's documented asynchronous API in supported regions. However, the official API reference currently marks Wan 3.0 as preview, so this article does not describe it as generally available.

03

Duration, resolution and native audio

Wan 3.0 supports up to 30 seconds in a single generation at 30fps when there is no video input. With a video input, the source plus generated output must remain within the documented 30-second limit.

Alibaba documents 480p, 720p and 1080p output. The model can generate dialogue, music and sound effects, and it supports editing plus forward, backward or bidirectional extension.

04

Images, video, audio, documents and web pages

Alibaba documents up to 10 reference images, five reference videos with a combined maximum of 15 seconds, and five reference audio files with a combined maximum of 15 seconds.

A request can also use one supported file or one public web link. Document formats include PDF, DOC/DOCX, PPT/PPTX, XLS/XLSX, text, Markdown and Apple productivity formats, within the current 100MB and 50-page limits.

First/last-frame control is mutually exclusive with the reference/file/link modes documented by Alibaba. The Wan 3.0 API guide maps these modes to a practical request flow.

05

Wan 3.0 pricing

For the Singapore region, Alibaba's current pricing page lists wan3.0-video at $0.05 per second for 480p, $0.10 for 720p and $0.20 for 1080p. It marks those rates as a limited-time 30% discount.

The same page lists wan3.0-video-prime at $0.068, $0.14 and $0.28 per second for 480p, 720p and 1080p. It also documents a conditional 30-second Singapore free quota. Region, tier and promotion terms can differ.

06

Is Wan 3.0 open source?

The reviewed official Wan 3.0 launch, API and pricing pages establish hosted preview access, but they do not establish an official Wan 3.0 open-weight release. Earlier Wan generations should not be used as evidence that Wan 3.0 weights or a license are available.

For a workflow comparison, see Wan 3.0 vs Gemini Omni 1.1 Flash vs Veo 3.1.

Sources

Primary and supporting sources

Facts were rechecked against the linked sources immediately before publication. Pricing, product availability and rollout status can change.

Project Monet

Useful signals. Clear decisions. Better digital work.

Project Monet turns relevant shifts in AI, creator tools and the web into practical context—and builds focused websites for businesses ready to grow.

Request a free homepage concept