Models / WAN
WAN AI Video Generator
WAN by Alibaba is a family of AI video models. The newest, WAN 3, came out in August 2026. Access WAN's latest video generation models for text-to-video, image-to-video with an end frame, reference-driven shots, and clips up to thirty seconds with audio, all within the same creative workflow.
Overview
What is WAN?
WAN is a video generation model built for longer clips, native audio, end-frame control, and reference-driven consistency. Access WAN models directly in AgentDocks and choose the right version for each shot.
Thirty Seconds in One Pass
Generate any length from two to thirty seconds with sound, without stitching clips together.
Start and End Frames
Give WAN 3 a first still and an optional last one, and the clip moves between them.
References by Number
Attach up to ten reference images and address each one in the prompt by its position.
Made on AgentDocks
What WAN Does Best
WAN stands out across real generation tasks, from thirty-second clips with audio to shots built from numbered reference images.
Holds a Scene for Thirty Seconds
Generates Sound with the Picture
Builds Shots From References
Also on the Canvas
WAN is One Menu Entry. Here are the Others.
