Supermicro Introduces 256 CPU NVIDIA Vera Rack for AI Agents

Oct 23, 2026
Supermicro introduces a liquid cooled rack with 256 NVIDIA Vera CPUs, support for up to 22,500 agent sandboxes, and security controls through BlueField-4 DPUs.
Supermicro Introduces 256 CPU NVIDIA Vera Rack for AI Agents

Supermicro announced in a press release a liquid cooled rack containing 256 NVIDIA Vera CPUs. The system can operate independently or alongside NVIDIA Vera Rubin NVL72 racks for agentic AI, reinforcement learning, and data processing workloads.

The rack provides 22,528 CPU cores and 45,056 threads, with support for up to 22,500 concurrent agent sandboxes. NVIDIA BlueField-4 DPUs handle networking and security functions, while Supermicro's DLC-2 system provides direct liquid cooling across the rack.

The system uses the third generation NVIDIA MGX design and supports either a conventional AC connection or an 800 VDC sidecar power rack with battery backup units. It supports NVIDIA OpenShell for runtime policy controls on the Vera CPUs and NVIDIA Sentry, which runs on the BlueField-4 DPUs as an out-of-band monitor. The rack is available for customer quotes now.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.