Google introduces Gemini Omni, a generative model for any-input creation, starting with video

Google introduced Gemini Omni, a generative model the company says combines Gemini's reasoning with creative generation, launching first as Gemini Omni Flash focused on video creation and editing. Google said the model accepts any combination of image, audio, video, and text inputs, supports conversational multi-turn editing with character consistency, and shows improved handling of physics such as gravity and fluid dynamics; all generated video carries SynthID watermarking.

The company said it is rolling out to Google AI Plus, Pro, and Ultra subscribers via the Gemini app and Flow, with free access on YouTube Shorts and YouTube Create, and API access for developers and enterprises to follow. Image and audio output modalities are planned for later, per Google; no benchmark figures were disclosed in the announcement.