Google releases Gemini 3 Pro with state-of-the-art reasoning and multimodal capabilities

Google has released Gemini 3 Pro in preview, introducing its most intelligent model to date with enhanced reasoning and multimodal understanding. The release includes a new Deep Think mode for complex problem-solving and integrates the model into Google Search, the Gemini app, AI Studio, Vertex AI, and the Google Antigravity platform.

• Tops the LMArena Leaderboard with a score of 1501 Elo.
• Achieves PhD-level reasoning scores on Humanity’s Last Exam (37.5%) and GPQA Diamond (91.9%).
• Sets new state-of-the-art benchmarks in mathematics (23.4% on MathArena Apex) and multimodal reasoning (81% on MMMU-Pro).
• Gemini 3 Deep Think mode improves performance on Humanity’s Last Exam to 41.0% and GPQA Diamond to 93.8%.
• Features a 1 million-token context window and supports zero-shot generation for richer web UIs.

The update aims to advance the state of the art in AI by providing deeper nuance, factual accuracy, and agentic capabilities across Google's ecosystem.