Google Introduces Gemini Omni: A New AI Model That Turns Any Input Into Video

Google has introduced Gemini Omni, a new multimodal AI model designed to take video generation and editing into a more advanced and accessible phase, with availability extending to users in the Arab world.

The announcement builds on Gemini’s previous progress in image generation and editing, particularly through Nano Banana, which helped millions of users restore old photos, create designs from simple sketches, and visualize ideas in ways that were not possible before.

With Gemini Omni, Google is now moving beyond image creation and into AI-powered video generation. The model can combine text, images, audio, and video as inputs, then generate high-quality videos grounded in Gemini’s understanding of the real world.

Video Creation From Any Type of Input

Gemini Omni brings together Gemini’s reasoning capabilities with a new layer of creative video generation. The model can understand multiple forms of input and turn them into a cohesive video output.

Instead of relying on complex editing tools, users can create and edit videos through natural conversation. They can change the environment, adjust character movement, add new objects, or completely reimagine a scene using simple prompts.

Google is rolling out the first model in the Omni family under the name Gemini Omni Flash, which will be available through the Gemini app, Google Flow, and YouTube Shorts. Over time, Google plans to expand Omni’s output capabilities to include other media formats such as images and audio.

Editing Videos Through Conversation

One of Gemini Omni’s most important features is the ability to edit videos using natural language. Each instruction builds on the previous one, allowing the model to preserve character consistency, realistic motion, and scene continuity.

Users can change specific details in a video, such as the background, camera angle, visual style, or individual elements. They can also start with an existing video and ask Omni to change what happens in the scene, add new characters, or transform an ordinary moment into something unexpected. This makes video editing feel more like a conversation and less like a technical process that requires advanced production tools.

A Deeper Understanding of Physics and Context

Gemini Omni is not only designed to create scenes that look realistic; it is also built to reason about what should happen next. The model draws on Gemini’s knowledge of physics, history, science, and cultural context to create more meaningful and coherent visual stories.

According to Google, Omni has an improved understanding of forces such as gravity, kinetic energy, and fluid dynamics, helping it generate more realistic motion and interactions within scenes.

The model can also turn complex ideas into visual explainers, making it useful for education, storytelling, content creation, and creative production.

Videos With a Personal Digital Avatar

Google is also working on advanced audio and speech editing capabilities for video, but says it is testing these features carefully before making them more broadly available.

As a first step, users will be able to create videos using their own AI avatar — a digital version of themselves that can look and sound like them. Google says this is part of its responsible approach to AI development, supported by clear policies designed to reduce misuse.

SynthID for AI Content Transparency

All videos created with Gemini Omni will include Google’s invisible digital watermark, SynthID, making it easier to verify whether a video was generated using AI.

Users will be able to verify Gemini Omni-generated videos through the Gemini app, Gemini in Chrome, and Google Search. This supports Google’s broader effort to improve content transparency and help people understand how digital content was created or edited online.

Availability

Gemini Omni Flash is rolling out now to subscribers of Google AI paid plans globally through the Gemini app and Google Flow. It is also being made available at no cost to users on YouTube Shorts and the YouTube Create app starting this week.

Google says the model will also become available to developers and enterprise customers through APIs in the coming weeks.

With Gemini Omni, Google is pushing AI video creation closer to a future where turning an idea into a full visual scene can happen through a simple conversation.