Nvidia CEO Jensen Huang calls AI titan’s latest chips ‘gigantic step up in performance’
Nvidia Unveils Next Generation Chips with Increased AI Computing Power
Nvidia CEO Jensen Huang announced at the Consumer Electronics Show in Las Vegas that the company’s next generation of chips is now in full production. These new chips are capable of delivering five times the artificial-intelligence computing power compared to Nvidia’s previous chips, especially when used for chatbots and other AI applications.
The new Vera Rubin platform, consisting of six separate Nvidia chips, is set to launch later this year. The flagship device will include 72 of the company’s top graphics units and 36 new central processors, with the ability to be combined into “pods” containing over 1,000 Rubin chips.
To achieve the enhanced performance, Huang mentioned that the Rubin chips utilize a proprietary data format that Nvidia hopes the industry will adopt in the future.
While Nvidia remains a dominant force in AI model training, it faces increasing competition from rivals like Advanced Micro Devices and customers such as Google in deploying AI technologies to a vast user base. Huang emphasized the new chips’ capabilities for tasks like improving chatbot responses through innovative storage technology.
Additionally, Nvidia introduced new networking switches with co-packaged optics connections and software for self-driving cars to make informed decisions and generate useful data for engineers. Huang also highlighted the company’s commitment to open-sourcing models and training data to build trust in AI technologies.
In recent developments, Nvidia acquired talent and chip technology from startup Groq to strengthen its position in the AI chip market. The company aims to showcase the superior performance of its latest products compared to older models like the H200, which has seen high demand, particularly in China.
