**Update:** Several users on social media have reported that the changelog has been modified to exclude references to Gemini 2. 0 Pro Experimental. This mention has also vanished for this reporter. A Google representative informed TechCrunch that an “out-of-date release note” was mistakenly “published. ” The original article continues: Google opted for a subtle approach in launching its next-gen flagship AI model, Gemini 2. 0 Pro Experimental. Instead of a grand reveal, the company introduced the model within a changelog for its Gemini chatbot application. The roll-out of Gemini 2. 0 Pro Experimental — the successor to last February’s Gemini 1. 5 Pro model — coincides with the tech industry’s keen focus on Chinese AI startup DeepSeek. DeepSeek’s recent models, freely available for companies to download and utilize, either match or exceed the performance of many top-tier models from American tech firms and AI organizations.

This has prompted a reckoning in Silicon Valley and among the upper echelons of the U. S. government. Starting Thursday, Gemini 2. 0 Pro Experimental is accessible to Gemini Advanced users and is now recognized as the leading model within Google’s Gemini AI lineup. The company states that it aims to deliver “better factuality” and “enhanced performance” in coding and mathematics tasks. “Whether you’re addressing advanced coding tasks, such as creating a specific program from scratch, or tackling mathematical problems like developing intricate statistical models or quantum algorithms, 2. 0 Pro Experimental will assist you in navigating even the most challenging tasks with improved ease and precision, ” Google mentions in the changelog. Gemini Advanced is part of Google’s Google One AI Premium subscription plan and is also available via Google’s Gemini for Google Workspace add-ons. Google indicates that Gemini 2. 0 Pro Experimental is currently in “early preview” status, which means it might exhibit “unexpected behaviors” and could generate errors. Additionally, unlike other Gemini models accessible through the app, Gemini 2. 0 Pro Experimental cannot access real-time data and is not compatible with certain app features. “We are committed to rapid iteration and delivering the best of Gemini to the world, and we want to ensure that Gemini Advanced subscribers have priority access to our latest AI advancements, ” Google added in the changelog. “Your feedback enables us to refine these models over time, and insights gained from experimental launches guide our wider release strategies. ”