NVIDIA and leading computer manufacturers have introduced a series of systems powered by the NVIDIA Blackwell architecture, featuring Grace CPUs, NVIDIA networking, and infrastructure. These systems are designed to facilitate the construction of AI factories and data centers aimed at advancing generative AI technologies.

During his keynote at COMPUTEX, NVIDIA founder and CEO Jensen Huang revealed that companies such as ASRock Rack, ASUS, GIGABYTE, Ingrasys, Inventec, Pegatron, QCT, Supermicro, Wistron, and Wiwynn will provide cloud-based, on-premises, embedded, and edge AI systems utilizing NVIDIA GPUs and networking solutions.

“The next industrial revolution has begun. Companies and countries are partnering with NVIDIA to shift the trillion-dollar traditional data centers to accelerated computing and build a new type of data center — AI factories — to produce a new commodity: artificial intelligence,” said Huang. “From server, networking and infrastructure manufacturers to software developers, the whole industry is gearing up for Blackwell to accelerate AI-powered innovation for every field.”

The product offerings will range from single to multi-GPUs configurations using x86- or Grace-based processors with air or liquid cooling technology. The NVIDIA MGX™ modular reference design platform now supports these Blackwell products. This includes the new NVIDIA GB200 NVL2 platform designed for large language model inference and data processing.

GB200 NVL2 is optimized for emerging market opportunities like data analytics. Utilizing high-bandwidth memory performance provided by NVLink®-C2C interconnects and dedicated decompression engines in the Blackwell architecture enhances data processing speeds by up to 18 times with eight times better energy efficiency compared to x86 CPUs.

NVIDIA MGX provides computer manufacturers with a reference architecture that allows them to build over 100 system design configurations efficiently. More than 90 systems from over 25 partners have been released or are in development using this architecture.

AMD and Intel are also supporting the MGX architecture with plans to deliver their own CPU host processor module designs. This includes AMD’s next-generation Turin platform and Intel® Xeon® 6 processor with P-cores (formerly Granite Rapids). These reference designs aim to reduce development time while ensuring consistency in design and performance.

The GB200 NVL2 leverages both MGX and Blackwell platforms enabling various system configurations that integrate seamlessly into existing data center infrastructures. It joins other products in the Blackwell lineup including NVIDIA Blackwell Tensor Core GPUs and GB200 Grace Blackwell Superchips.

NVIDIA’s partner ecosystem includes TSMC as well as global electronics makers providing essential components for creating AI factories. These include innovations from companies such as Amphenol, Asia Vital Components (AVC), Cooler Master, Colder Products Company (CPC), Danfoss, Delta Electronics, LITEON among others.

New data center infrastructure can be rapidly developed using these components alongside technologies like NVIDIA Quantum-2 or Quantum-X800 InfiniBand networking and BlueField®-3 DPUs from major system makers like Dell Technologies, Hewlett Packard Enterprise (HPE), Lenovo among others.

Enterprises can access the NVIDIA AI Enterprise software platform which includes NIM™ inference microservices for creating production-grade generative AI applications.

Taiwanese companies are also adopting Blackwell technology extensively. Chang Gung Memorial Hospital plans to use it for biomedical research while Foxconn aims to develop smart solution platforms for electric vehicles (EVs) and robotics using Grace Blackwell technology.

Supporting quotes from various industry leaders highlight their commitment towards integrating advanced interconnect solutions; efficient cooling; tailored server lineups; innovative connector technology; thermal management solutions; quick disconnect fluid power designs; advanced cooling systems; support across GPUs/CPUs/DPUs/high-speed networking among other areas critical for enabling accelerated computing capabilities required by modern enterprises globally.

To learn more about these developments watch Jensen Huang’s COMPUTEX keynote presentation.

___