Google Unveils Gemini 3, Touts PhD-Level Reasoning

News Desk
Islamabad: Google has launched Gemini 3, calling it the company’s “most intelligent model” yet, marking a major leap in its race for dominance in the rapidly evolving AI landscape.
CEO Sundar Pichai announced the rollout, describing Gemini 3 as a flagship system capable of state-of-the-art reasoning and designed to understand depth and nuance in complex problems. The top-tier model, Gemini 3 Pro, is available immediately in the Gemini app and — for the first time — integrated into Google Search from day one.
Developers can access the model through AI Studio and Vertex AI. Google says Gemini 3 significantly outperforms its predecessor, Gemini 2.5 Pro, posting leading scores on major benchmarks, including PhD-level reasoning and advanced mathematics.

Read More: https://thepenpk.com/cloudflare-bug-takes-chunk-of-web-offline/
A key upgrade in the new model is its shift away from “flattery” and “cliché,” with Gemini 3 engineered to deliver direct, concise, and authentic insights, acting more like a “thought partner” than a conversational assistant.
The model also supports interactive visualizations for learning complex subjects, improved long-horizon planning for multi-step tasks, and enhanced app development through what Google calls “vibe coding.”
Alongside the Pro version, Google introduced Gemini 3 Deep Think, a boosted reasoning model designed for even more rigorous problem-solving.
For developers, the company unveiled Google Antigravity, a new agentic development platform powered by Gemini 3 that can autonomously plan and execute advanced software tasks — signaling Google’s push toward AI-driven automation.
Google also emphasized safety, noting Gemini 3 has undergone extensive security and risk evaluations, making it the company’s most secure model to date.
The rollout begins immediately, marking a significant milestone in Google’s accelerating AI roadmap.