Google was already making headlines with Nano Banana, the image generation model that has generated over 50 billion images to date. This year, Google has leveled up with a new Gemini Omni model, which the search giant claims “can create anything from any input — starting with video.”
In other words, the new model lets you combine images, audio, video, and text as input to generate high-quality videos and edit them through conversational prompts. Gemini Omni Flash is the first model in the family, and Google will add support for other output modalities, such as images and audio, in the future.
During the Google I/O 2026 keynote, the search giant described several things you can do with the new model. While editing videos, your characters, scene sequences, and other aspects stay consistent, and you can use natural language prompts to add or remove objects, pick different camera angles, or change them entirely.
You can feed a real-life video you shot into Omni and ask it to add new characters or objects, or sprinkle in some CGI effects. “Change the environment, angle, style or even specific details, without ever losing the thread of your original scene. Scroll through the carousel to see how edits build on each other,” Google explained.
The search giant notes that Omni has an “intuitive understanding of physics,” and its creations are grounded in Gemini’s knowledge about history, science, and cultural context to improve photorealism and storytelling. The AI model can generate explainer videos from short prompts, complete with AI voiceovers.
You can use images of characters, scenes, or drawings as input. While it supports multiple input formats, Google said the model will initially support only voice references for audio, with other audio types added later.
Note that all videos created using Omni will have the SynthID watermark, which you can verify through the Gemini app, Chrome, and Google Search. You can also create an avatar or digital version of yourself that can feature in the videos generated by Omni, looking and sounding like you.
Gemini Omni Flash is now rolling out to the Gemini app and Google Flow for users with a Google AI Plus/Pro/Ultra subscription. It’s also rolling out for free to users who want to remix YouTube Shorts and in the YouTube Create app.
Omni has arrived alongside the new Gemini 3.5 Flash model, which is Google’s “strongest” coding model yet. You can check out more coverage from Google I/O 2026.
