Google launches Gemini Omni Flash, a conversational video-generation model with avatar mode held back
Summary
Google DeepMind unveiled Gemini Omni Flash, the first Gemini Omni model for conversational video generation and editing from mixed image/audio/video/text inputs. It’s rolling out to the Gemini app and Google Flow for Google AI Plus/Pro/Ultra subscribers and free in YouTube Shorts/Create, with developer and enterprise API access coming. Flash supports multi-turn edits that preserve scene and character continuity, claims improved handling of physical interactions, defaults to SynthID watermarking, caps clips at 10 seconds, and withholds general-purpose speech/audio editing while offering an avatar mode that requires user-recorded onboarding.
Why it matters
This product rollout brings high-quality multimodal video generation into mainstream apps, accelerating adoption while making provenance, consent, and moderation immediate operational priorities.