Google has introduced Gemini 2.0, the latest upgrade in its AI technology, designed to push the boundaries of usability and innovation. Building on Gemini 1.0 and 1.5, this release brings advanced multimodal abilities and reasoning power to deliver a smarter, more versatile AI experience.

The model supports both multimodal input and output, processing text, images, video, and audio while generating seamless responses across these formats. New features include steerable text-to-speech in multiple languages and native image generation. Its advanced reasoning capabilities make it adept at solving complex problems, handling intricate queries, and even coding.

The experimental Gemini 2.0 Flash model offers low-latency performance while surpassing earlier versions in speed and benchmarks. It also integrates with tools like Google Search and third-party applications, giving developers even greater flexibility.

In Google Search, Gemini 2.0 will enhance AI-powered overviews, enabling it to address complex questions and advanced tasks. The new "Deep Research" feature adds further value, acting as an AI research assistant capable of exploring topics and creating detailed reports.

Built on Google’s Trillium TPUs, Gemini 2.0 represents a decade of innovation in AI infrastructure. Now available to developers and trusted testers, it sets the stage for a future where AI serves as a proactive and universal assistant.