ChipsSep 24, 2026 · 00:172 sources

Supermicro Ships Nvidia Rubin AI Racks for Massive Deployments

From marketbrief.now

Image: marketbrief.now
AI;DR

Supermicro began shipping Nvidia-powered Vera Rubin NVL72 racks on September 23, with each rack containing 72 Rubin GPUs, 36 Vera CPUs, and 20.7 TB of HBM4 memory. The systems provide 216 TB/s via NVLink for AI model training and inference, and Supermicro is also offering a 16 rack Scalable Unit featuring 1,152 Rubin GPUs and 331 TB of HBM4 with a direct liquid cooling stack for data center deployments ranging from 5 MW to gigawatt scale.

Read the sourceSign in to suggest

Sources

#nvidia#chips#infra#hbmMarkdown
Content history (2)
  • 2026-10-04 · Summary · vi · Replaced stored text
    Mỗi rack Vera Rubin NVL72 chứa 72 GPU Rubin, 36 CPU Vera và 20,7 TB bộ nhớ HBM4. Hệ thống…
    Mỗi rack Vera Rubin NVL72 chứa 72 Rubin GPU, 36 Vera CPU và 20,7 TB bộ nhớ HBM4. Các hệ t…
  • 2026-10-04 · Title · vi · Replaced stored text
    Supermicro bắt đầu xuất xưởng rack AI Nvidia Rubin phục vụ triển khai quy mô lớn
    Supermicro giao các rack AI Nvidia Rubin cho các hệ thống triển khai quy mô lớn

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today