Google recently announced Gemini, an AI model designed to meet various demands, from lightweight mobile applications to heavyweight enterprise solutions. Gemini comes in three versions: Nano, Pro, and Ultra, each tailored for specific purposes and areas of use.
Nano is the lightweight version of the Gemini AI model, specifically crafted for native and offline use on Android devices. It empowers mobile applications with on-device AI capabilities, ensuring powerful performance, quick response times, and enhanced privacy. Users of the Pixel 8 Pro will benefit from Nano's new features, highlighting its practical application in enhancing the user experience on mobile devices.
A six-minute video showcases Gemini's multimodal capabilities, demonstrating tasks such as speech recognition paired with image identification. Gemini quickly identifies images for Gemini users, even connect-the-dot images, within seconds. It tracks pen strokes and ball movements in real-time, showcasing its capabilities. While humans can perform these tasks, Gemini excels at understanding what's next and making predictions—it's an AI.
According to Google CEO Sundar Pichai, Gemini is poised to integrate into Google's ecosystem alongside various aspects like the Google Search Engine, advertising products, Chrome browser, and more, positioning Gemini as part of Google's future plans. While the model suite is currently available only in English, Google aims to soon support other languages.
Google's CEO of DeepMind, Demis Hassabis, mentions that among 32 benchmark tests, Gemini outperformed ChatGPT in 30 of them in evaluating AI-based capabilities. While Gemini won most benchmark tests, it excelled particularly in understanding audio and video inputs even better."