Right now at its I/O convention, Google unveiled Gemini 3.5, its newest household of AI fashions, together with Gemini Omni, a brand new mannequin that may create video from any enter.
The primary obtainable mannequin within the Gemini 3.5 household is Gemini 3.5 Flash. That is now out for everybody through the Gemini app and it is also obtainable in AI Mode in Google Search. This mannequin “delivers intelligence that rivals giant flagship fashions on a number of dimensions, on the speeds you could have come to anticipate from the Flash sequence”, Google says.
It is the strongest agentic and coding Gemini mannequin, outperforming even Gemini 3.1 Professional on difficult coding and agentic benchmarks, whereas main in multimodal understanding. Gemini 3.5 Flash is now the default mannequin.
Gemini Omni is a brand new mannequin that may create movies from any enter. You possibly can mix photos, audio, video, and textual content as enter and it’ll generate high-quality movies “grounded in Gemini’s real-world data”. As soon as a video is created, it could possibly additionally simply be edited via dialog.
The primary mannequin within the Gemini Omni household is Gemini Omni Flash, which helps you to change particular issues a few video, or all the things, and you may refine the creations throughout a number of turns, altering issues with out ever dropping the thread of the unique scene.
The mannequin has “an improved intuitive understanding of forces like gravity, kinetic power and fluid dynamics”, so it could possibly create extra reasonable scenes. With Omni you need to use your personal voice and Avatars, which create a digital model of your self. The entire movies embrace SynthID digital watermarking.
Gemini Omni Flash is out there at present for all subscribers to Google AI Plus, Professional, and Extremely plans globally within the Gemini app and in Google Stream. It is also rolling out without cost to customers on YouTube Shorts and YouTube Create.