"The new GPU will work "hand-in-hand" with the Vera CPU and be housed inside the latest Nvidia Vera Rubin NVL 144 CPX platform, the company said. This is an integrated NVIDIA MGX system that packs 8 exaflops of AI compute to provide 7.5x more AI performance than NVIDIA's GB300 NVL72 systems. It also promises 100TB of fast memory and 1.7 petabytes per second of memory bandwidth in a single rack."
""The Vera Rubin platform will mark another leap in the frontier of AI computing - introducing both the next-generation Rubin GPU and a new category of processors called CPX," said Jensen Huang, founder and CEO of Nvidia. "Just as RTX revolutionized graphics and physical AI, Rubin CPX is the first CUDA GPU purpose-built for massive-context AI, where models reason across millions of tokens of knowledge at once.""
Nvidia's Rubin series introduces a new class of GPU purpose-built for massive-context processing such as extensive software coding and high-speed video generation. The Rubin GPU is designed to operate alongside the Vera CPU within the Vera Rubin NVL 144 CPX platform, an integrated NVIDIA MGX system delivering 8 exaflops and 7.5x the AI performance of GB300 NVL72 systems. The platform promises 100TB of fast memory and 1.7 petabytes per second of memory bandwidth per rack. Rubin CPX will be offered in flexible configurations, targets general release in 2026, and uses a cost-efficient monolithic die with NVFP4 resources optimized for high-performance, energy-efficient inference and long-context processing.
Read at IT Pro
Unable to calculate read time
Collection
[
|
...
]