Gemini AI
Gemini AI
Google has announced the release of Gemini 2.5, its latest AI model. The company says this new version is its most intelligent AI to date, with the first release being an experimental version of 2.5 Pro.
Gemini 2.5 has "thinking models," capable of reasoning through their thoughts before responding. This results in better performance and improved accuracy, according to Google. The company explains that in AI, "reasoning" refers to a system's ability to analyze information, draw logical conclusions, incorporate context and nuance, and make informed decisions.
As per Google, Gemini 2.5 Pro tops the LMArena leaderboard by a big margin, which Google says means it's a highly capable model with high-quality style. It also shows strong reasoning and code capabilities.
Google Gemini 2.5 Pro
Google Gemini 2.5 Pro
Google Gemini 2.5 Pro Credit: Devesh Beri
Gemini 2.5 Pro is now available in Google AI Studio and in the Gemini app for Gemini Advanced users. It will be coming to Vertex AI soon, with pricing details to be announced in the coming weeks.
Google revealed that Gemini 2.5 Pro performs well on a range of benchmarks that require new reasoning. It scores 18.8% across models without tool use on Humanity's Last Exam, a dataset made to capture the human frontier of knowledge and reasoning.
In terms of coding, Google says Gemini 2.5 represents a big leap over 2.0. The new version excels at creating visually effective web apps and agentic code applications, along with code transformation and editing.
It has a 1 million token context window, with a 2 million token window coming soon. Google says this helps the model understand large amounts of data and solve complicated problems using different types of information, such as text, audio, images, video, and even entire code repositories.