Alibaba Releases Wan3.0 Video Generation Model With Native Audio and 30-Second Clips

Alibaba has released Wan3.0, a major update to its open-source AI video generation model, capable of producing 30-second cinematic videos with synchronized audio in a single pass. The model, which entered public beta on August 6 and launched widely on August 24, represents a significant leap over its predecessor—doubling the maximum video length from 15 to 30 seconds while maintaining up to 1080p resolution and introducing native dialogue, sound effects, and background music generation.