Google launches Gemini 3.1 Pro, doubling reasoning performance and reclaiming top AI model spot

Google released Gemini 3.1 Pro, which scores 77.1% on ARC-AGI-2, more than doubling Gemini 3 Pro's 31.1% and outperforming Anthropic's Opus 4.6 (68.8%) and OpenAI's GPT-5.2 (52.9%). It also hit 94.3% on GPQA Diamond, 2887 Elo on LiveCodeBench Pro, and 80.6% on SWE-Bench Verified. The model introduces three-tier thinking levels (low, medium, high), where high functions as a "mini Deep Think." Third-party evaluator Artificial Analysis confirmed it's now the most powerful AI model available. It's rolling out across Google AI Studio, Gemini CLI, Vertex AI, Gemini Enterprise, Android Studio, the Gemini app, and NotebookLM.

View full digest for February 20, 2026