Google DeepMind Launches Gemini Omni Flash with Video Generation

Google DeepMind has introduced Gemini Omni Flash, the first model in the Omni family, capable of generating and editing videos from any combination of inputs including text, images, audio, and video. The model builds on last year's Nano Banana image generation and editing capabilities. Omni Flash uses Gemini's multimodal reasoning to understand physics, cultural context, and real-world knowledge, enabling features like consistent character editing, physics-accurate scenes, and explainer videos. Users can edit videos through natural language conversation, with edits building on previous instructions. The model supports input references for style, motion, or character transfer. All generated videos include SynthID digital watermarking. The rollout began today to Google AI Plus, Pro, and Ultra subscribers globally via the Gemini app and Google Flow, and at no cost to YouTube Shorts and YouTube Create App users starting this week. Developer and enterprise API access will follow in coming weeks. Audio output modalities are planned for future release.