Googles CEO, Sundar Pichai has introduced the Gemini era, a phase, in AI development that he believes will bring about a transformation in the field. Gemini is a language model in different versions tailored to specific needs. For instance there’s Gemini Nano, which is designed to be lightweight and capable of offline functioning on Android devices. Then we have Gemini Pro, a version powering Google AI services and serving as the foundation for Bard. Lastly there’s Gemini Ultra, a model specifically built for data centers and enterprise applications.
Google is currently integrating Gemini into its systems. Bard already benefits from the capabilities of Gemini Pro while users of Pixel 8 Pro can enjoy the functionalities provided by Gemini Nano. Starting from December 13th developers and enterprise customers can access Gemini Pro through Google Generative AI Studio or Vertex AI on Google Cloud. Although it currently supports English language processing only there are plans to expand Geminis language support in the future. Pichai envisions a future where Gemini plays a role in enhancing Googles search engine functionality, ad products, Chrome browser experience and more.
Both Sundar Pichai and DeepMind CEO Demis Hassabis have highlighted Gemnis performance as they pose it as a challenge, to OpenAIs ChatGPT model.
In order to showcase its superiority, over models like ChatGPT Google extensively evaluated the capabilities of its model in tasks, including multi task language understanding and Python code generation. They conducted 32 benchmarks to assess its performance.
According to Hassabis Google outperformed OpenAI in 30 out of the 32 benchmarks demonstrating the performance of their model. This indicates that Google is responding to OpenAIs prominence in the AI field and their rapid technological advancements.
The introduction of Gemini and its capabilities is a milestone for Google as they aim to solidify their position in the AI landscape. By focusing on adaptability and efficiency, across applications Gemini aims to shape the direction of Googles AI endeavors.
