Wan 3.0 Reference to Video

Lock in exactly what matters — a face, a product, a motion, a voice — instead of describing it from scratch. Drop in up to 10 images, 5 video clips, and 5 audio clips, then point at them in your prompt ("the character from Image 1 picks up the object in Image 3"), and Wan 3.0 assembles it into a single take. Generates up to 30 seconds in one run with audio produced alongside the picture, in 480P, 720P, or 1080P. Alibaba's latest video model — ready to run on Nordy, no install, no GPU.

• Video & Animation
• API
• Image to Video
• Wan