Google Unveils Gemini 3: Enhanced Reasoning, Multimodal Features, and Seamless Search Integration

Google has introduced Gemini 3, its most advanced AI model yet, boasting significant improvements in reasoning and multimodal understanding (text, images, videos, and code). With over 650 million monthly users and widespread adoption among Cloud customers, Gemini 3 Pro reportedly excels in AI benchmarks, offering capabilities such as processing inputs of up to one million tokens and enhanced mathematics and video understanding. Notably, it integrates directly into Google Search to optimize query comprehension and generate dynamic visual responses. Available to Google AI Pro and Ultra subscribers in the U.S., Gemini 3 supports developers through platforms like Google AI Studio and Vertex AI, promoting innovations like “vibe-coding” for interactive app creation. While the technology shows promise, questions remain about real-world reliability and potential biases in AI outputs. Google emphasizes responsible development amidst rapid growth in AI investment, as it navigates the evolving landscape of consumer-focused AI experiences.

Source link