
Genjutsu AI generates videos by transferring motion from a reference clip and swapping objects with user‑provided photos. Users select a short source video, choose a preset or model such as Seedance or Kling, and preview a credit estimate before rendering the final output.