Supermicro Strengthens U.S. Manufacturing and AI Solutions with NVIDIA Collaboration

Supermicro unveiled its plans to deliver next-generation NVIDIA AI platforms, specifically the NVIDIA Vera Rubin NVL144 and NVIDIA Vera Rubin NVL144 CPX, expected to launch in 2026. These platforms will enhance AI training and inference capabilities, empowering organizations to efficiently handle complex AI workloads. The announcement comes as part of Supermicro's focused strategy to cater to federal customers with strict compliance and performance standards.

The company also introduced a range of U.S.-manufactured systems compliant with the Trade Agreements Act (TAA). Notable among these is the high-density 2OU NVIDIA HGX B300 8-GPU system, which can support up to 144 GPUs per rack. This powerful system is designed to meet the escalating demand for AI and HPC deployments in government data centers.

Supermicro CEO Charles Liang emphasized the company's robust commitment to domestic manufacturing during the press conference. "Our expanded collaboration with NVIDIA and our focus on U.S.-based manufacturing position Supermicro as a trusted partner for federal AI deployments," Liang stated. The company’s operations, including headquarters, manufacturing, and R&D, are based in San Jose, California, which provides a strategic advantage in delivering high-quality technology solutions to American federal customers.

The company’s rigorous validation process for its government-optimized systems ensures compliance with federal regulations and enhances supply chain security. This focus on local production is seen as a key differentiator in the competitive landscape of IT solutions.

One of the highlights from Supermicro's press release is the introduction of the new liquid-cooled ARS-511GD-NB-LCC Super AI Station. This innovative platform integrates the high-end NVIDIA GB300 Superchip into a compact desktop form factor, offering over 5x AI PFLOPS of computing power compared to traditional workstations. This solution is ideal for government agencies, startups, and research labs requiring high-performance AI model training and development in a more accessible format.

Supermicro also announced the general availability of its ARS-121GL-NB2B-LCC NVL4 rack-scale platform. This solution is designed for GPU-accelerated HPC and AI workloads across a variety of scientific fields, including molecular simulation and genomics. The platform achieves revolutionary performance through the integration of four NVIDIA NVLink-connected Blackwell GPUs and two NVIDIA Grace CPUs, demonstrating Supermicro's commitment to scalable and high-performance solutions.

In addition to hardware innovations, Supermicro is set to support the newly announced NVIDIA BlueField-4 DPU and NVIDIA ConnectX-9 SuperNIC technologies. These advancements will enable faster AI networking and data processing capabilities, further enhancing Supermicro's offerings for AI infrastructure. The company’s modular hardware design will facilitate the seamless integration of these new technologies, ensuring rapid deployment and cost-effective development.

Supermicro's press release on October 28, 2025, illustrates the company’s dedication to advancing U.S.-manufactured AI infrastructure solutions. By expanding its collaboration with NVIDIA and focusing on compliance and performance for federal applications, Supermicro positions itself as a key player in the rapidly evolving landscape of AI technology. With a strong commitment to local production and innovative product offerings, the company is poised to meet the growing demands of government agencies and other sectors looking to harness the power of AI.