Google’s new Gemini Pro model has record benchmark scores — again

Google Unveils Gemini 3.1 Pro: The Latest Breakthrough in AI Technology

Google has recently introduced the newest version of Gemini Pro, known as Gemini 3.1. This advanced Language Model (LLM) is currently available as a preview and is set to be officially released soon, marking a significant milestone in the realm of artificial intelligence.

Gemini 3.1 Pro has garnered attention as one of the most powerful LLMs to date. Building upon the success of its predecessor, Gemini 3, which was already highly regarded upon its launch in November, this latest model represents a substantial leap forward in AI capabilities.

Independent benchmarks, such as Humanity’s Last Exam, have showcased the superior performance of Gemini 3.1 Pro compared to its previous iteration. The model’s enhanced functionality and efficiency have been acknowledged by industry experts and technology enthusiasts alike.

Brendan Foody, CEO of AI startup Mercor, commended Gemini 3.1 Pro for its exceptional performance on APEX, a benchmarking system designed to evaluate AI models’ proficiency in real-world tasks. Foody stated in a social media post that the model has secured the top spot on the APEX-Agents leaderboard, underscoring its prowess in handling complex knowledge work.

The release of Gemini 3.1 Pro coincides with a competitive landscape in the AI industry, as companies vie to introduce cutting-edge LLMs tailored for agentic functions and multi-step reasoning. Google’s unveiling of this advanced model reinforces its commitment to advancing AI technology and staying ahead in the AI model wars.