Model · Alibaba (Tongyi Lab) · Proprietary

Wan3.0

Text-to-videoImage-to-videoVideo editProprietary
Generate Wan3.0 on AIContentDrop →
Arena rating
No votes yet
Crowd votes (Design Arena board)
3,891
Arena status
Pre-launch
Facts checked
2026-10-06

Vendor-sourced facts

Published facts for Wan3.0 — vendor-sourced, checked 2026-10-06
FactValue
VendorAlibaba (Tongyi Lab) [1]
LicenceProprietary (API/app only) [1]
Licence detailAPI only via Alibaba Cloud Model Studio (wan3.0-video, wan3.0-video-prime) and wan.video; weights not published (last open-weight Wan release is Wan2.2) [1]
ModesText-to-video, Image-to-video, Video edit [1]
Release2026-08-24 [1]
Max resolution1080P [1]
Max clip duration30 s [1]
Native audioYes [1]
AccessAPI, web app [1]
Vendor pagehttps://www.alibabacloud.com/help/en/model-studio/use-video-generation [1]

Arena rating: no votes yet. Nothing on this page is a score, a rank or a measurement by us — every value is what Alibaba (Tongyi Lab) publishes, with the page it came from listed under Sources.

Crowd votes · external board

On Design Arena's video board Wan3.0 appears as “Wan 3.0”: Elo 1329 · #4 of 48 · win rate 63.5% ± 1.5% · 3,891 votes (2,469W 1,422L) · avg 219.5s.[2] These are their voters' numbers, not this arena's — our own arena has no votes yet.

Source: Design Arena (designarena.ai/leaderboard/video), checked 2026-10-21.

About Wan3.0

Wan3.0 is Alibaba's August 2026 model, served from Alibaba Cloud Model Studio and wan.video. It is widely described as open, and that is the single most common error about it: Alibaba's own Hugging Face organisation publishes weights only up to the Wan2.2 family, so Wan3.0 is catalogued here as proprietary, API-only. [1]

The stated ceilings are 1080P and thirty seconds with native audio, and Alibaba notes the model can be called without an application process. [1] Among the API-only entries it is one of the few that combines a long clip ceiling with general availability rather than a waitlist, which makes it a natural opponent for the open-weights models in a battle.

“now officially available and can be called without the need to apply”

Alibaba (Tongyi Lab) on Wan3.0 [1]

Want hands-on context before trusting anyone's ranking? Generate your own clips with a multi-model workflow on AIContentDrop: price a long-clip workflow · fire the job at the REST API.

Sources

  1. Alibaba (Tongyi Lab) — Wan3.0, vendor documentation https://www.alibabacloud.com/en/news/product/video-generation-model-wan30-officially-launched-nsb — checked 2026-10-06
  2. Design Arena — video leaderboard https://www.designarena.ai/leaderboard/video — checked 2026-10-21

Each [n] above is the vendor page the fact beside the marker was read from, on the date shown. Where a vendor states nothing, the table says “Not stated” and no value is inferred from anywhere else.