Google’s latest AI model, Gemini 3 Pro, has reclaimed top positions on AI benchmarks, outperforming previous models in the LMArena and WebDev Arena leaderboards. It enhances Google Search with an innovative AI Mode, providing users with smarter, multimodal responses that combine text and visuals. This advanced model features in Google Antigravity, an integrated development environment that enables advanced agentic coding capabilities. Gemini 3 Pro delivers better reasoning and creativity, achieving impressive scores, including 1,501 on the LMArena. The rollout also introduces Gemini 3 Deep Think for Google AI Ultra subscribers, promising even superior reasoning performance. Now integrated into Google AI Studio, Vertex AI, and third-party platforms like GitHub, Gemini 3 fosters independent coding through AI agents that execute tasks autonomously. Available through the Gemini app and Google Search, Gemini 3 is designed to surface relevant information effectively, transforming user interactions. Experience Gemini 3 and its groundbreaking features today.
Source link
