Scenes from several references
Combine characters, places and objects from different photos in one scene.
ALIBABA · IN COCUYO
Alibaba’s video model that combines text, frames and references in a single generation, with audio. Wan 3 Prime is its fast mode.
Try Wan 3Explore for free. Create with an account and credits.

AT A GLANCE
WHAT TO CREATE
Combine characters, places and objects from different photos in one scene.
From 2 seconds for a gesture to 30 for a full scene.
Test with Wan 3 Prime and move to Wan 3 once you have the shot.
TRY THIS PROMPT
A boy and his dog run through a sunflower field towards the camera, a yellow kite in the sky, wind moving the flowers, cheerful background music, wide shot ending in a close-up.
Each version opens with its own controls, references and credit cost.
CREATE AT YOUR PACE
Buy a top-up or choose an optional plan. Check the cost of each creation before you confirm.
Yes, Wan 3 can generate the clip with audio. Check the model controls in the Studio before confirming.
The same Wan 3 in fast mode, designed for iterating.
Up to 10 images, 5 videos and 5 audio clips, or start and end frames.