Google has launched an updated preview of its advanced AI model, Gemini 2.5 Pro, just two months after its previous version. This iteration enhances performance in programming tasks and excels in coding benchmarks, math, science, knowledge, and reasoning. Developers can access it via Google AI platforms like AI Studio and Vertex AI. Notably, it features improved response style and structure, delivering more creative and well-formatted answers. The model now supports up to 2 million tokens—double the previous capacity—and integrates user feedback for refinement. CEO Sundar Pichai highlighted its advancements in coding and reasoning capabilities. The pricing is set at $1.25 per million input tokens and $10 per million output tokens. Emphasizing the development of “thinking models,” Google aims for more accurate and efficient responses. Overall, Gemini 2.5 signifies a significant leap in performance through enhanced foundational models and refined techniques.
Source link
Google Enhances Gemini 2.5 Pro: CEO Sundar Pichai Unveils Its Creative and Intelligent Upgrades

Leave a Comment
Leave a Comment