Skills
Gemini Omni Flash: Natural-Language Iterative Video Editing — Describe Changes, Keep Everything Else Intact, Ship 10-Second Clips
Google's Gemini Omni Flash generates video from any combination of text, image, audio, and video inputs with physics-aware rendering. Key differentiator over competitors: iterative editing via natural language. Describe what to change ('shift camera left', 'make sculpture from bubbles', 'when person touches mirror, make it ripple like liquid') and Omni reworks the specific element while preserving everything else. Rolling out through Gemini, Flow, YouTube Shorts, and YouTube Create with API access planned.
Source
↳ Follow the thread