--- format: "aidr-story-markdown/v1" id: "a990aeb9356464ebb20456dc3c18519428a4893489ddee7e14c762e1aa520b3e" canonical_url: "https://aidr.today/a990aeb9?lang=en" title: "Supermicro Ships Nvidia Rubin AI Racks for Massive Deployments" lang: "en" requested_lang: "en" available_langs: ["en","vi"] translation_fallback: null fallback_fields: [] published_at: "2026-09-23T17:17:53.000Z" category: "Chips" topics: ["nvidia","chips","infra","hbm"] source_urls: ["https://marketbrief.now/ai/supermicro-ships-nvidia-rubin-ai-racks-for-massive-deployments-713abccb","https://huggingnews.com/ai/supermicro-ships-nvidia-rubin-ai-racks-for-massive-deployments-713abccb"] summary: "Each individual Vera Rubin NVL72 rack contains 72 Rubin GPUs, 36 Vera CPUs, and 20.7 TB of HBM4 memory. The systems are designed for high scale-up bandwidth, providing 216 TB/s via NVLink to facilitate faster AI model training and inference. Supermicro began shipping these Nvidia-powered units on September 23 and is also offering a 16 rack Scalable Unit featuring 1,152 Rubin GPUs and 331 TB of HBM4. This larger configuration includes a direct liquid cooling stack intended for data center deployments ranging from 5 MW to gigawatt scale." --- # Supermicro Ships Nvidia Rubin AI Racks for Massive Deployments > [Open the canonical story]() **Published:** 2026-09-23T17:17:53.000Z **Category:** Chips **Topics:** nvidia, chips, infra, hbm ## Summary Each individual Vera Rubin NVL72 rack contains 72 Rubin GPUs, 36 Vera CPUs, and 20\.7 TB of HBM4 memory\. The systems are designed for high scale\-up bandwidth, providing 216 TB/s via NVLink to facilitate faster AI model training and inference\. Supermicro began shipping these Nvidia\-powered units on September 23 and is also offering a 16 rack Scalable Unit featuring 1,152 Rubin GPUs and 331 TB of HBM4\. This larger configuration includes a direct liquid cooling stack intended for data center deployments ranging from 5 MW to gigawatt scale\. ## Sources - [Story source]() - [Story source]()