“Visual Layout creates an immersive, magazine-style look with photos and modules,” says Josh Woodward, VP of Google Labs, Gemini, and AI Studio. “These elements don’t just look good, but invite your input to further develop the results.”
Along with Gemini 3, Google is also introducing Gemini Agent, an experimental feature designed to handle multi-step tasks directly within an app. The agent can connect to services such as Google Calendar, Gmail, and Reminders. Once accessed, it can perform tasks such as managing the inbox or managing the schedule.
Like other agents, it breaks tasks into discrete steps, displays its progress in real time, and pauses for approval from the user before continuing. Google describes this feature as a step towards “a true generalist agent”. It will be available on the web for Google AI Ultra users in the US starting November 18.
The overall approach resembles “vibe coding” where users define an end goal in plain language and let the model assemble the interface or code needed to get there.
The update integrates Gemini more deeply into Google’s existing products. In Search, a limited group of Google AI Pro and Ultra users can now switch to Gemini 3 Pro, in new Model Reasoning modes, to get deeper, more complete AI-infused summaries that rely on model reasoning instead of existing AI modes.