Alibaba Launches Wan3.0 Video Model, Accepts Document Inputs

Alibaba Cloud has officially launched Wan3.0, a video-generation model designed to create video clips up to 30 seconds in length. The model expands its input capabilities to include various document formats such as DOC, XLS, PPT, PDF, and Markdown.

The model has been undergoing public beta testing since early August. Alibaba Cloud stated that ongoing development has focused on enhancing instruction following, improving consistency across shots, and refining audio quality.

This release aims to streamline video creation for businesses and content creators by allowing them to leverage existing documents for video generation. Alibaba Cloud is offering a temporary 30% discount on Wan3.0 API pricing for selected platforms as part of the launch.