Google launches Gemini 3 with new programming app and record results

Google on Tuesday launched Gemini 3, its latest and most advanced core model, which is now instantly available through the Gemini app and AI search interface.

This new model comes just seven months after the release of Gemini 2.5, Google’s most capable LLM program to date, and a direct competitor to the most capable AI tool on the market. The release also comes less than a week after OpenAI released GPT 5.1, and just two months after Anthropic released Sonnet 4.5 — a reminder of the rapid pace of frontier model development.

A more research-intensive version of the model, called Gemini 3 Deepthink, will also be made available to Google AI Ultra subscribers in the coming weeks, once it passes further rounds of safety testing.

“With Gemini 3, we’re seeing this huge leap in thinking,” said Tulsi Doshi, Google’s product lead for the Gemini model. “She responds with a level of depth and nuance we’ve never seen before.”

Some of this logical power has already been captured in independent standards. With a score of 37.4, the model scored the highest score ever recorded on the recent Test of Humanity test, which aims to capture general thinking and experience. The previous high score obtained by GPT-5 Pro was 31.64. Gemini 3 also topped the leaderboard in LMArena, a human-driven benchmark that measures user satisfaction.

According to Google, Gemini currently has more than 650 million monthly active users, and 13 million software developers have used this model as part of their workflow.

Alongside the basic model, Google has also released a Gemini-powered encryption interface called Google Antigravity, allowing multi-part proxy encryption similar to IDEs like Warp or Cursor 2.0. Specifically, Antigravity combines a ChatGPT-style prompt window, a command-line interface, and a browser window that can show the effect of changes made by the coding agent.

“The agent can work with your editor, across your terminal, and across your browser to make sure it helps you build this application in the best way possible,” said Koray Cavukoglu, chief technology officer at DeepMind.