Hardware
How the NVIDIA Vera Rubin Platform is Solving Agentic AI’s Scale-Up Problem
The NVIDIA Vera Rubin Platform is a rack-scale AI supercomputer designed to power agentic AI and reasoning models at scale by eliminating bottlenecks in communication and memory movement for efficient
The NVIDIA Vera Rubin Platform is a rack-scale AI supercomputer designed to power agentic AI and reasoning models at scale by eliminating bottlenecks in communication and memory movement for efficient multi-step workflows. The platform delivers more tokens per watt and lower cost per token compared to the NVIDIA Blackwell architecture generation. The Vera Rubin NVL72 combines 72 Rubin GPUs and 36 Vera CPUs, training mixture-of-experts models with one-fourth the GPUs required by Blackwell while delivering up to 10x higher inference throughput per watt and one-tenth the cost per token.
Source: NVIDIA Developer | 2026-05-14