Google has introduced Gemini Omni, a new multimodal AI model designed for video creation and editing using combinations of text, images, video, and audio inputs.
The first model in the Omni family, Gemini Omni Flash, is being rolled out to the Gemini app, Google Flow, and YouTube Shorts.
According to the company, the model allows users to edit videos through conversational prompts, with each instruction building on previous edits while maintaining continuity across scenes, characters, and visual elements.
The tool also supports video generation using multiple forms of input references, including images, drawings, video clips, and voice inputs. Google said support for additional audio input formats will be added later.
The company stated that Gemini Omni combines Gemini’s reasoning capabilities with video generation functions, including scene creation informed by concepts such as physics, historical context, and visual consistency.
Google also introduced an avatar feature that allows users to create videos using a digital version of themselves and their own voice.
All videos generated using Gemini Omni will include SynthID digital watermarks, according to the company. Google said generated content can also be verified through the Gemini app, Gemini in Chrome, and Google Search.
The launch expands Google’s broader Gemini ecosystem, which previously included image-generation and editing tools such as Nano Banana.
