Nvidia Vera Rubin NVL72 rack
Also written: NVL72
Nvidia's Vera Rubin NVL72 is a dense server rack for AI, packing 72 GPUs for extreme computing power.
The Nvidia Vera Rubin NVL72 rack is a specialized, high-power computer system. It's designed to hold many powerful graphics processing units (GPUs) from Nvidia, which are essential for training and running advanced AI models.
Think of it like a super-powered server cabinet. Instead of just a few computers, it packs 72 GPUs into a single unit, allowing for massive parallel processing. This is crucial for the huge calculations needed in AI development, like teaching a computer to understand language or generate images.
The Nvidia NVL72 is a high-density, liquid-cooled server rack designed to house 72 Nvidia H100 or H200 GPUs. It's optimized for AI training and inference workloads, offering significant compute power and memory bandwidth in a compact form factor. The system integrates networking and power delivery for these GPUs, aiming to simplify deployment of large-scale AI infrastructure.
Key features include its dense GPU configuration and integrated cooling solutions, which are necessary to manage the heat generated by such a high concentration of powerful processors. This allows for greater compute density compared to traditional server setups, reducing the physical footprint and potentially improving power efficiency per GPU.
This term is appearing now due to the increasing demand for massive computing power to train and deploy frontier AI models. Companies like xAI are exploring partnerships for advanced hardware solutions, including specialized racks like the Nvidia NVL72, to meet these growing computational needs for their AI development.
- The SpaceX-Nvidia partnership heads for orbitxAI · The Rundown AI ·
- xAI Grok
- xAI is partnering with Nvidia, indicating a potential use or interest in Nvidia's advanced hardware solutions like the Vera Rubin NVL72 rack for their AI development.
- Nvidia DGX H100
- The DGX H100 is a server system containing 8 H100 GPUs, whereas
The Nvidia Vera Rubin NVL72 rack is a product developed by Nvidia, a leading technology company specializing in graphics processing units and AI hardware. It is named after Vera Rubin, a pioneering astronomer. Source ↗
While you can't directly interact with an NVL72 rack, you can explore Nvidia's AI Enterprise software. This software suite provides tools and frameworks optimized to run on Nvidia hardware, allowing you to experiment with AI models that would typically run on such powerful systems.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- What is the main purpose of the Nvidia Vera Rubin NVL72 rack?
- Its main purpose is to provide a highly dense and powerful computing platform for training and running large-scale artificial intelligence models and other high-performance computing tasks.
- How many GPUs can the NVL72 rack hold?
- The NVL72 rack is designed to hold up to 72 Nvidia H100 or H200 GPUs.
First seen on Subvolts Aug 25, 2026. Updated Sep 15, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.