Alibaba Cloud said on August 24 that Alibaba’s video generation model Wan3.0 had launched.

The model can generate 30-second videos in a single run and, for the first time, supports document inputs, including doc, xls, ppt, pdf, and md files.

The company said Wan3.0 was upgraded in generation duration, creation flexibility, reference capabilities, and real-world restoration.

The announcement did not provide performance data.

The release comes as China’s rules require labeling of AI-generated or manipulated audio and video. Online deep-synthesis service providers may need to add visible or embedded markers, retain records, and, in some cases, require real-name registration.

Chinese social media platforms have also added labels and tightened content review to comply, suggesting similar governance features may matter for large-scale AI video services delivered through cloud platforms.

🔗 Source: Yicai

Recent Alibaba Cloud developments