Google is launching Gemini 3, its most intelligent AI model yet. From the outset, users will have access to the flagship Gemini 3 Pro model, which is multimodal from the ground up and can process text, image, audio, and video in the same flow. This model is said to top several AI benchmark tests in logic, math, and fact checking. It’s also said to provide shorter, more direct, and less biased responses.
One new feature is that Gemini 3 can generate interfaces in the form of presentations or web page layouts based on the user’s instructions. There is also an improved search feature with extended “query fan-out” that will allow the model to better understand the intent behind a question.
In both the Gemini app and Google Search’s AI Mode, answers are now also displayed in a way that combines visual elements such as images, tables, grids, and even interactive simulations.
On top of that, Gemini 3 Pro has gained sharper reasoning and planning abilities, allowing it to handle more long-term and complex tasks. It powers Google’s new experimental tool Gemini Agent, which can, for example, organize emails or book travel for users.
Gemini 3 Pro is now available in the Gemini app in preview form. In the US, Google AI Pro and Ultra subscribers can also test the Agent feature as well as the new Advanced AI mode.