Google is pushing deeper into AI-generated content material with Gemini Omni, a brand new mannequin designed to show totally different types of enter into editable movies by pure language directions.
Introduced at Google I/O 2026, the transfer expands Gemini past textual content and picture era into multimodal video creation. Gemini Omni combines Gemini’s reasoning capabilities with generative instruments to create video outputs from textual content, photos, audio and video inputs.
The primary mannequin within the household, Gemini Omni Flash, is rolling out by the Gemini app, Google Move and YouTube Shorts. Within the coming months, help for extra output codecs together with photos and audio can also be anticipated.
Additionally Learn: Google offers Gemini its personal 24/7 AI assistant with Gmail and Workspace integrations
From prompts to edits by dialog
In contrast to typical AI video instruments that always require repeated prompts and separate enhancing steps, Gemini Omni is constructed round conversational enhancing. Customers can proceed refining movies throughout a number of directions whereas sustaining continuity from earlier modifications.
Characters stay constant throughout scenes, edits retain context from earlier prompts and movies will be modified with out restarting the artistic course of. Customers can alter environments, change actions, add objects or introduce totally new components whereas preserving the move of the unique scene.
The system additionally goals to deliver higher realism to generated content material by making use of a broader understanding of physics and contextual information.
Combining a number of inputs into one video
Gemini Omni can work throughout a number of types of media concurrently. Current movies, photos, sketches and audio information can be utilized as references and reworked right into a single output.
The mannequin additionally attracts on Gemini’s broader understanding of historical past, science and cultural context to create explainers and visible storytelling codecs alongside artistic content material era.
Google additionally launched avatar options that permit customers to create digital variations of themselves utilizing their very own voice for AI-generated movies.
Gemini Omni Flash is rolling out globally to Google AI Plus, Professional and Extremely subscribers and also will turn into accessible on YouTube Shorts and YouTube Create.
First Revealed on