Vidu Q4 is an AI video generation model for turning a single starting image, or a prompt with reference pictures, into short video clips.

What it does

• Image to video: animate one starting image into a short clip.
• Reference to video: combine a prompt with at least one reference picture.
• Choose motion, video length and resolution before rendering.

Where it fits Storyboards, product teasers and social cutdowns where a still frame needs to move. Teams that already produce key visuals can reuse them instead of shooting new footage.

The product page describes the workflow as: pick a starting image or a prompt plus references, then set motion, length and resolution, then render. Longer edits still need an NLE, so treat the output as source material rather than a finished cut.

More details and the live demo: Vidu Q4