Google Unveils Gemini AI Model That Will Power Bard AI

Google has been consistently enhancing its AI chatbot, the Google Bard, with new features. However, it has faced tough competition from ChatGPT and other AI chatbots utilizing GPT-3 and GPT-4. In a recent development, Google has unveiled that the Bard is now powered by Gemini Pro, a brand-new AI model from Google.

Described by Google as its most capable AI model to date, Gemini is designed to operate seamlessly across various systems, from data centers to mobile devices. Its versatility lies in its ability to process and integrate text, code, audio, images, and videos.

Google’s Gemini 1.0 boasts multimodal capabilities, enabling it to comprehend complex information and answer related questions. It is available in three sizes: Gemini Ultra for demanding tasks, Gemini Pro for versatile applications, and Gemini Nano for streamlined on-device operations.

The integration of Gemini Pro into Google Bard represents a significant upgrade, empowering the chatbot with advanced reasoning, planning, and understanding capabilities. This enhancement is set to be accessible in English across more than 170 countries and territories. The Pixel 8 Pro will soon receive an update to run Gemini Nano, bringing features like Summarize in the Recorder app, Smart Reply in Gboard, and WhatsApp integration.

Gemini Ultra, according to Google, excels in a broad spectrum of tasks, including natural image, audio, and video understanding, mathematical reasoning, and multimodal problem-solving. It has surpassed human experts on the MMLU benchmark, evaluating both knowledge and problem-solving abilities, and achieved an outstanding score on MMMU, a multimodal reasoning benchmark. Notably, Ultra has outperformed previous models on image benchmarks without relying on object character recognition, showcasing its native multimodality and complex reasoning capabilities.

Google plans to conduct early testing and gather feedback from a select group of users for Gemini Ultra before a broader rollout to developers and enterprise customers in early 2024. Additionally, Bard Advanced, an AI experience powered by Gemini Ultra, is set to launch in early 2024, providing users with access to cutting-edge capabilities.

In addition to these advancements, Google has introduced Cloud TPU v5p, the most powerful TPU system to date. This system enables faster training of large-scale generative AI models and accelerates the development of new products and capabilities. Gemini employs dedicated safety classifiers to identify, label, and filter out harmful content, ensuring safety and inclusivity for all users through a layered approach.

Follow INCPAK on Google News / Facebook / Twitter / Instagram for updates.