Supermicro Computer, Inc. (SMCI), a Total IT Solution Provider for AI/ML, HPC, Cloud, Storage, and 5G/Edge, has officially confirmed plans to deliver next-generation NVIDIA AI platforms, including the NVIDIA Vera Rubin NVL144 and NVIDIA Vera Rubin NVL144 CPX in 2026.
According to certain reports, the company also took this opportunity to launch U.S.-manufactured, TAA (Trade Agreements Act)-compliant systems. More on the given systems would reveal how they include the high-density 2OU NVIDIA HGX B300 8-GPU system with up to 144 GPUs per rack, as well as an expanded portfolio featuring a Super AI Station based on NVIDIA GB300 and the new rack-scale NVIDIA GB200 NVL4 HPC solutions.
Talk about Supermicro’s latest slew of solutions on a slightly deeper level, we begin from how the company plans on expanding its portfolio of solutions based on NVIDIA HGX B300 and B200, NVIDIA GB300 and GB200, and NVIDIA RTX PRO™ 6000 Blackwell Server Edition GPUs. For better understanding, each of these systems arrive on the scene bearing an ability to provide unprecedented compute performance, efficiency, and scalability for key federal government use cases, such as cybersecurity & risk detection, engineering & design, healthcare & life sciences, data analytics & fusion platforms, modeling & simulation, and secure virtualized infrastructure.
in case that wasn’t enough, all these government-optimized systems are also developed, constructed, and rigorously validated at its global headquarters in San Jose, California, where the company ensures full compliance with the TAA and eligibility under the Buy American Act.
Building upon that, Supermicro has further made a decision to expand its government-focused portfolio through NVIDIA AI Factory for Government reference design, which happens to be a full-stack, end-to-end reference design, capable of providing guidance for deployment and management of multiple AI workloads on-premises and in the hybrid cloud.
Complementing that would be the introduction of 2OU NVIDIA HGX B300 8-GPU server. This particular solution features an OCP-based rack-scale design powered by Supermicro Data Center Building Block Solutions®. The idea behind conceiving such a mechanism is rooted in supporting up to 144 GPUs in a single rack, while simultaneously delivering outstanding performance and scalability for large-scale AI and HPC deployments in government data centers.
Another detail worth a mention relates to Supermicro relaying support for the newly-announced NVIDIA BlueField-4 DPU and NVIDIA ConnectX-9 SuperNIC in gigascale AI factories. You see, these new accelerated infrastructure technologies will be ready, right at the moment of launch, for an integration into new Supermicro AI systems to provide faster cluster-scale AI networking, storage access, and data processing offload for the next generation of NVIDIA AI infrastructure.
To make the proposition even more attractive, the company’s modular hardware design should also come in handy here to facilitate rapid integration of new technologies, such as the NVIDIA BlueField-4 and NVIDIA ConnectX-9 into existing systems designs with minimal re-engineering, thus speeding up time-to-market and reducing development costs.
Supermicro even marked its latest announcement with the release of its new liquid-cooled ARS-511GD-NB-LCC Super AI Station. This station, on its part, brings the high-end server grade GB300 Superchip into a deskside form factor, all for achieving more than 5x AI PFLOPS of computing power, as compared to traditional PCIe based GPU workstations.
This new Super AI Station also makes up a complete solution for AI model training, fine-tuning, applications, as well as algorithms prototyping and development, a solution which can be deployed on-prem for unmatched latency and full data security, supporting models up to 1 trillion parameters. In essence, such a proponent can prove to be ideal for government agencies, startups, deep-tech and research labs who may not have access to traditional server infrastructure for AI development purposes.
It can also aid all those organizations that are unable to leverage cluster-scale or cloud AI services because of availability, cost, privacy, and latency concerns.
“Our expanded collaboration with NVIDIA and our focus on U.S.-based manufacturing position Supermicro as a trusted partner for federal AI deployments. With our corporate headquarters, manufacturing, and R&D all based in San Jose, California, in the heart of Silicon Valley, we have an unparalleled ability and capacity to deliver first-to-market solutions are developed, constructed, validated (and manufactured) for American federal customers,” said Charles Liang, president and CEO, Supermicro.
