Tired of video editing? Google’s Gemini Omni changes scenes when you ask

Google unveiled Gemini Omni at I/O, a multimodal model that can generate and edit coherent videos from text, images, audio, and video.

The model keeps context across multi-step edits, aiming for smoother continuity, more natural conversational editing, and mixed-media workflows.

Google will watermark Gemini Omni outputs with SynthID to help identify AI-generated media and address misuse concerns.

Gemini Omni Flash is rolling out first to select subscribers and products, with enterprise API access planned later.