Gemini 3 Deep Think Upgrade: 84.6% on ARC-AGI-2, 48.4% on Humanity’s Last Exam
Google’s Gemini 3 Deep Think just received a major upgrade.This time, the focus is not on flashy demos but on serious, research-grade performance. In partnership with scientists and researchers, Deep Think is now tackling some of the toughest reasoning benchmarks ever created. And the numbers are raising eyebrows in the AI community. What Is Gemini … Read more