Alibaba's Wan3.0 video generation model, now available in beta, doubles the maximum length of AI-generated videos to 30 seconds in a single pass. The most significant update for developers and content teams is the ability to accept static documents — PDFs, PowerPoint presentations, spreadsheets, and web pages — as direct input for video creation.

Add AlexTech.ai asPreferred Source on Google

From static documents to dynamic clips

Wan3.0 processes text, images, video, and audio simultaneously: a single prompt can include up to 10 images, 5 video clips, and 5 audio files. According to Alibaba, the model aims to fix common visual drift and distortion issues in AI videos by keeping characters, props, and spatial layouts consistent with reference materials. The system also includes a tool to extend existing videos and automatically suggests the optimal length based on the prompt.

Alibaba launches Wan3.0 video model with 30-second generation and ... — https://news.lavx.hu/article/alibaba-launches-wan3-0-video-model-with-30-second-generation-and-document-input

Pricing and availability

The model is accessible via wan.video, Alibaba Cloud Model Studio, or through the API on Qwen Cloud. Two tiers are available: a Standard version, currently at a 30% discount, and a faster Prime version. A 30-second clip in 1080p costs $6.00 with Standard and $8.40 with Prime; in 720p, prices drop to $3.00 and $4.20, while 480p starts at $1.50.

The financial context behind the launch

Wan3.0's release comes at a critical moment for Alibaba's AI strategy: the company just completed the largest share sale among Hong Kong-listed companies to fund its AI push. The previous week, Alibaba reported a 75% year-over-year drop in quarterly profit, attributed to sharply higher AI investments. The model fits into a recent series of releases from Tongyi Lab, including Qwen-Image-3.0 for graphic generation and Qwen-Audio-3.0 for speech synthesis, consolidating the company's multimodal approach.

Toward professional production and robotics training

Alibaba positions Wan3.0 for a wide range of applications: from film production to short drama and social media clips, up to training autonomous vehicles and robotics systems using realistic simulated footage. With competition intensifying — from ByteDance's Seedance 2.5 to MiniMax H3 — the ability to turn corporate documents into dynamic videos represents a concrete competitive advantage for enterprise workflows.