Supermicro Introduces the NVIDIA Vera CPU Rack, a 256 CPU, 100% Liquid-Cooled Rack-Scale Solution, Complementing the NVIDIA Vera Rubin Platform, for Agentic AI at Scale

• Delivers dedicated 256 NVIDIA Vera CPU capacity for agent sandboxes, complementing NVIDIA Vera Rubin NVL72 with common DCBBS infrastructure with third-generation NVIDIA MGX racks and battery backup units
• Integrates NVIDIA Vera CPUs and BlueField™-4 DPUs with the NVIDIA Open Agent Safety Platform and OpenShell Reference Architecture for secure, high-density agentic AI orchestration and continuous monitoring and policy enforcement
• Leverages Supermicro DCBBS to deliver a complete, integrated data center solution—from liquid-cooled rack architecture and power infrastructure to networking, cooling, and deployment services—enabling customers to build and scale full AI factories for agentic workloads

SAN JOSE, Calif., Oct. 22, 2026 /PRNewswire/ -- Super Micro Computer, Inc (NASDAQ: SMCI), an AI, Enterprise, Storage, and 5G/Edge Total IT Solution Provider featuring Data Center Building Block Solutions® (DCBBS), today announced the NVIDIA Vera CPU rack as the latest addition to the DCBBS portfolio. The NVIDIA Vera CPU rack delivers dense CPU capacity for agentic AI, reinforcement learning (RL) workloads, and data processing, supporting 22.5K sandboxes at liquid-cooled AI factory scale. Complementary to the now shipping Supermicro NVIDIA Vera Rubin NVL72, the 100% liquid-cooled CPU rack can be used either standalone or alongside NVIDIA Vera Rubin NVL72 in AI factories, providing clusters with high-density CPU compute, designed to address the rapidly escalating requirements of agentic AI.

"As agentic AI transforms data center requirements, Supermicro is extending its DCBBS portfolio with fully integrated infrastructure," said Charles Liang, president and CEO of Supermicro. "For customers seeking air-cooled deployments, Supermicro's NVIDIA Vera CPU portfolio includes air-cooled 1U and 2U systems, in addition to rack-scale liquid cooling options. Together with NVIDIA, we help accelerate time to deployment with a common purpose-built architecture for secure, scalable agentic AI."

To watch a video summary of this release, click here, and for the latest information on Supermicro solutions featuring NVIDIA Vera Rubin, click here.

"As enterprises scale agentic AI and reinforcement learning, supporting thousands of concurrent agent workflows requires a new class of high-density CPU infrastructure," said Chris Marriott, vice president of enterprise platforms at NVIDIA. "NVIDIA Vera CPUs and BlueField-4 DPUs, integrated with Supermicro's liquid-cooled DCBBS architecture, give customers the dedicated compute and hardware-based security to scale the execution of agentic workloads across their AI factories."

Built to power the age of Agentic AI

Agentic AI and RL workloads represent an evolution from linear inference chains to branched, and often looped, execution with many more intermediate steps that depend on CPU availability and performance. GPUs remain the primary engine for generating tokens, code, and queries, while CPU-based sandbox environments execute actions, evaluate results, run tools, and orchestrate workflows. At the scale of modern AI factories, with hundreds or thousands of deployed GPUs, dedicated CPU compute is needed to host thousands or millions of parallel execution environments for agents.

The NVIDIA Vera CPU rack from Supermicro integrates 256 NVIDIA Vera CPUs, delivering 22,528 cores and 45,056 threads and supporting up to 22,500 concurrent agent sandboxes per rack. NVIDIA BlueField-4 DPUs provide high-performance networking and in-silicon security, and Supermicro's DLC-2 direct liquid cooling stack enables a 100% liquid-cooled rack built for these workloads.

Common Infrastructure with Third-Generation MGX Racks and Battery Backup Units

The Vera CPU rack uses the third-generation NVIDIA MGX design, part of the DLC-2 stack, creating a common infrastructure architecture with Vera Rubin NVL72 that supports standardized deployment. NVIDIA MGX is a modular reference architecture for accelerated computing, helping system builders design future-compatible systems faster, reducing costs, accelerating time to market, and improving ROI. To power the compute racks, customers can choose either a traditional AC feed to compute rack or an 800 VDC sidecar power rack, which enables the use of battery backup units (BBU) in place of a conventional UPS for power loss protection. This approach helps free data center floor space, can improve end-to-end power conversion efficiency, and lowers CapEx and OpEx.

Prioritizing Secure Agentic AI with OpenShell Reference Architecture

The NVIDIA Vera CPU rack from Supermicro natively supports NVIDIA OpenShell Reference Architecture to continuously monitor and govern agent behavior, extending Supermicro's support for secure AI agents. NVIDIA OpenShell runs on NVIDIA Vera CPUs to apply least-privilege, policy-based controls to agents at runtime, while NVIDIA Sentry runs on NVIDIA BlueField-4 DPUs as an out-of-band monitor that enforces policy, provides attested telemetry, and can quarantine an agent that moves outside its permitted boundaries in milliseconds. Together, these software solutions can help companies retain observability and auditability in production agentic systems without consuming host CPU resources.

An Optional Companion to Vera Rubin NVL72

The Vera CPU rack is designed to work alongside Vera Rubin NVL72 and is equally suited to standalone deployment. For customers building on the Supermicro DCBBS Blueprints for NVIDIA Vera Rubin NVL72, Supermicro offers an optional design that pairs Vera Rubin NVL72 racks with Vera CPU racks, with a defined network topology built on NVIDIA Spectrum-X™ Ethernet, hot-aisle containment layouts, and redundant in-row coolant distribution units (CDUs) and rear door heat exchangers from the DLC-2 stack. A single Supermicro team supports site survey, design, L11 and L12 integration and testing, on-site deployment, and ongoing support, whether the Vera CPU rack is deployed with Vera Rubin NVL72 or on its own.

Flexible Deployment Beyond NVIDIA Vera Rubin NVL72

The Vera CPU rack also operates as a standalone CPU platform. Standalone deployments can accelerate agent orchestration and sandboxing, database operations, cluster management, and schedulers such as Slurm and Kubernetes, data analytics, and large-scale software build pipelines.

Availability

The Supermicro NVIDIA Vera CPU rack is available to quote now. Learn more about Supermicro's NVIDIA Vera CPU rack, and other solutions in the NVIDIA Vera Rubin platform here.

About Super Micro Computer, Inc.

Supermicro (NASDAQ: SMCI) is a global leader in Application-Optimized Total IT Solutions. Founded and operating in San Jose, California, Supermicro is committed to delivering first-to-market innovation for Enterprise, Cloud, AI, and 5G Telco/Edge IT Infrastructure. We are a Total IT Solutions provider with server, AI, storage, IoT, switch systems, software, and support services. Supermicro's motherboard, power, and chassis design expertise further enables our development and production, enabling next-generation innovation from cloud to edge for our global customers. Our products are designed and manufactured in-house (in the US, Taiwan, and the Netherlands), leveraging global operations for scale and efficiency and optimized to improve TCO and reduce environmental impact (Green Computing). The award-winning portfolio of Server Building Block Solutions® allows customers to optimize for their exact workload and application by selecting from a broad family of systems built from our flexible and reusable building blocks that support a comprehensive set of form factors, processors, memory, GPUs, storage, networking, power, and cooling solutions (air-conditioned, free air cooling or liquid cooling).

Supermicro, Server Building Block Solutions, and We Keep IT Green are trademarks and/or registered trademarks of Super Micro Computer, Inc.

All other brands, names, and trademarks are the property of their respective owners.

SOURCE Super Micro Computer, Inc.