Google has launched Gemini, its most advanced and capable artificial intelligence model to date, as the tech giant seeks to enhance its AI offerings and explore new monetization strategies.
Gemini, which includes three different sizes – Gemini Ultra, Gemini Pro, and Gemini Nano – will be available for licensing to customers through Google Cloud for use in their own applications.
Gemini (Google’s Largest AI Model)
Gemini Ultra, the largest and most capable category of the model, is designed to excel in a wide range of tasks, while Gemini Pro scales across various tasks, and Gemini Nano is optimized for specific tasks and mobile devices.
Starting December 13, developers and enterprise customers can access Gemini Pro via the Gemini API in Google AI Studio or Google Cloud Vertex AI.
Android developers will also have the opportunity to build with Gemini Nano.
Gemini will power Google products like its Bard chatbot and the Search Generative Experience (SGE), which aims to answer search queries with conversational-style text.
Although SGE is not widely available yet, it represents a significant step forward in Google’s search technology.
Gemini Ultra has outperformed human experts on the Massive Multitask Language Understanding (MMLU) test, which assesses world knowledge and problem-solving abilities across 57 subjects, including math, physics, history, law, medicine, and ethics.
This achievement highlights Gemini Ultra’s ability to understand nuance and reasoning in complex subjects.
Google CEO Sundar Pichai emphasized that Gemini was built to be multimodal, meaning it can understand, operate across, and combine different types of information, including text, code, audio, image, and video.
This approach allows for more versatile and comprehensive AI applications.
Bard’s Integration with Gemini
Google’s chatbot Bard will use Gemini Pro to enhance its reasoning, planning, understanding, and other capabilities.
Early next year, Google plans to launch “Bard Advanced,” which will utilize Gemini Ultra, representing the most significant update to the chatbot.
With all being Google’s largest AI model, Gemini Ultra is noted for being significantly cheaper to serve, highlighting its efficiency.
Google’s ability to train these models has become more efficient, offering more capabilities at a lower cost.
Google plans to release a technical white paper with more details about Gemini, although it will not disclose the perimeter count.
The company’s next-generation tensor processing unit, TPU v5p, will also support training AI models, offering improved performance for the price compared to the TPU v4.
Gemini represents a major advancement in Google’s AI technology, with potential applications across various fields and industries.
This new era of models is one of the most significant science and engineering efforts undertaken by Google, promising to unlock new opportunities for people everywhere.
