Recently, Google officially launched its new core model Gemini 3.1 Pro, marking a new stage in the breakthrough of artificial intelligence technology.
Gemini 3.1 Pro is specifically designed for complex problems in scientific, engineering, and research fields, with a focus on enhancing core reasoning capabilities, achieving significant improvements in efficiency and accuracy in solving cutting-edge challenges. Official information shows that the model performed excellently in multiple rigorous benchmark tests. For example, in the ARC-AGI-2 test, which evaluates logical pattern processing capabilities, the actual performance of Gemini 3.1 Pro reached 77.1%, with reasoning performance more than doubling compared to the previous generation Gemini 3 Pro.
To promote the widespread adoption of the technology, Google has also launched a multi-channel distribution plan. Developers can experience the preview version through Google AI Studio, Gemini CLI, the agent development platform Antigravity, and the Gemini API in Android Studio. Enterprise users can access the services by logging into Vertex AI and Gemini Enterprise platforms; ordinary consumers can directly use it in the Gemini app and NotebookLM. It is worth mentioning that Google AI Pro and Ultra subscription users will enjoy higher usage limits, and NotebookLM will exclusively open up the full functionality of 3.1 Pro for these two types of users.
