Super Micro Computer, Inc. unveiled its upcoming system portfolio powered by the NVIDIA Vera Rubin platform. The company positioned the new lineup around changing data center demands as facilities evolve into Artificial Intelligence factories built to produce intelligence at massive scale.
Supermicro said agentic reasoning, long-context Artificial Intelligence, and Mixture-of-Experts (MoE) workloads are driving demand for a new class of compute and storage infrastructure. Its NVIDIA Vera Rubin NVL72, NVIDIA HGX Rubin NVL8, and NVIDIA Vera CPU systems are being designed and built with Supermicro’s Data Center Building Block Solutions (DCBBS) advanced liquid-cooling technology stack to accelerate time-to-market for customers.
The company framed the announcement around rising inference requirements and the need for infrastructure that can support next-generation deployments. Charles Liang, president and CEO of Supermicro, said organizations increasingly require an Artificial Intelligence factory to compete, and he described the DCBBS technology stack as a way to give customers a fast, clear path to deploying upcoming NVIDIA Vera Rubin NVL72, HGX Rubin NVL8, and Vera CPU systems at scale.
Supermicro also presented the new systems as an early look at infrastructure intended for the next phase of Artificial Intelligence development. The announcement emphasized the company’s goal of reaching market quickly with platforms designed for large-scale data center deployments and advanced cooling needs.
