Nvidia’s groundbreaking Grace Hopper Superchips Are in Full Production Now

Nvidia has kicked off the production of the Nvidia GH200 Grace Hopper Superchip. This chip is engineered to support systems running sophisticated AI applications.

The systems powered by the GH200, featuring over 400 configurations based on Nvidia’s latest CPU and GPU architectures, including Grace, Hopper, and Ada Lovelace, are primarily aimed at high-performance computing (HPC) tasks. They are set to be available for purchase later this year. latest During the Computex event in Taiwan, Nvidia's CEO Jensen Huang showcased new systems and partners while sharing further details regarding the GH200 Grace Hopper Superchip. This chip synergizes the Arm-based Nvidia Grace CPU and Hopper GPU alongside Nvidia's NVLink-C2C interconnect technology.

With an impressive total bandwidth exceeding 900GB/s, the GH200 achieves bandwidth capabilities that are up to seven times greater than traditional PCIe Gen5 lanes found in standard accelerated systems, offering extraordinary computational power for demanding generative AI and HPC applications.

"Generative AI is swiftly reshaping industries, revealing new possibilities and speeding up advancements in sectors like healthcare, finance, business services, and beyond. Now that the Grace Hopper Superchips are in full-scale production, manufacturers worldwide are gearing up to create the enhanced infrastructure that enterprises require to develop and launch generative AI applications tailored to their unique proprietary data,\" stated Ian Buck, Nvidia's vice president of accelerated computing.

Prominent hyperscalers and supercomputing centers located in Europe and the U.S. are among the entities that will have access to systems powered by the GH200. Notable global system manufacturers, including Aeon, Advantech, Aetina, ASRock Rack, Asus, Gigabyte, Ingrasys, Inventec, Pegatron, QCT, Tyan, Wistron, and Wiwynn, are adopting this advanced technology.

Leading global server producers, such as Cisco, Dell Technologies, Hewlett Packard Enterprise, Lenovo, Supermicro, and Eviden (an Atos company), are providing a diverse range of Nvidia-accelerated systems. In addition, prominent cloud partners for Nvidia H100 encompass Amazon Web Services, Cirrascale, CoreWeave, Google Cloud, Lambda, Microsoft Azure, Oracle Cloud Infrastructure, Paperspace, and Vultr. Nvidia AI Enterprise, a software framework integral to the Nvidia AI platform, offers over 100 frameworks, pretrained models, and development tools, including solutions for generative AI, computer vision, and speech AI.

Olaplex, a well-known haircare brand, has just made its entrance into the Decentraland metaverse through the launch of a virtual influencer.

Read more related articles:

• DeepMind has introduced a new AI tool known as Dramatron, which has the capability to generate comprehensive drafts for movies or television scripts.
• Meta has established a new product team that concentrates on generative AI, while Elon Musk is assembling a team with the goal of creating a competitor to ChatGPT.
• Please be aware that the content on this page is not designed to serve as legal, tax, investment, financial, or any other form of professional advice. It's crucial to only invest what you can afford to lose and to seek independent financial guidance if you have concerns. For additional details, we encourage you to refer to the terms and conditions as well as the help and support sections provided by the issuer or advertiser. MetaversePost strives for accurate and unbiased reporting, yet market conditions may change without prior notice.