9 Demo Cases of Gemini Omni and Gemini 3.5

At Google I/O 2026, the next-generation Gemini Omni and Gemini 3.5 model families were announced.

Gemini Omni is a model that generates high-quality content from a variety of inputs, including video. It can combine images, audio, video, and text to produce videos, and videos can be edited through natural language conversation alone. In particular, it excels at maintaining character consistency, adhering to the laws of physics, and remembering the context of previous scenes.

Gemini 3.5 Flash is a model focused on building powerful AI agents by combining state-of-the-art intelligence with execution capability. Through Antigravity technology, it carries out complex, long-step workflows, and it delivers excellent performance on practical tasks such as automatic classification of unstructured assets and coding work.

In addition, 3.5 Flash is being applied as the default model for the Gemini app and Google Search. In particular, the Information agents introduced into Search operate in the background 24 hours a day, proactively finding information users need and providing comprehensive updates.