Nvidia says its new DGX GH200 AI supercomputer offers the power needed to build the next generation of large-scale generative AI models. Powered by GH200 Grace Hopper Superchips and the NVLink Switch System, the DGX GH200 is expected to be available to customers by the end of the year. Google Cloud, Meta, and Microsoft are among the first companies to gain access to the DGX GH200. - The supercomputer can support the development of advanced language models for AI chatbots, recommendation algorithms, fraud detection and data analysis, and more.
- It combines 256 GH200 Superchips, providing 1 exaflop of performance and 144 terabytes of shared memory.
- According to Nvidia, it eliminates the need for CPU-to-GPU PCIe connections and offers increased bandwidth and power efficiency.
- Nvidia is also building its own AI supercomputer, named Helios, that's comprised of four DGX GH200 systems and 1,024 Grace Hopper Superchips.
Nvidia also revealed ACE for Games, an AI-powered system for enhancing video game character development. |
