<p>➀ NVIDIA launched the Rubin CPX GPU, a specialized accelerator for massive-context AI models, delivering 30 PetaFLOPS of NVFP4 performance and 128 GB of GDDR7 memory on a monolithic die;</p><p>➁ The GPU is optimized for disaggregated inference, separating compute-bound context phases and memory bandwidth-bound generation phases to enhance throughput, reduce latency, and improve resource utilization;</p><p>➂ Integrated with NVIDIA Vera CPUs and Rubin GPUs in the Vera Rubin NVL144 CPX platform, it provides 8 exaflops of AI compute, 7.5x faster than previous systems, and scales to 100TB of memory and 1.7PB/s memory bandwidth per rack.</p>
Related Articles
- AI Gigafactory Bollox3 months ago
- Onward and upward for Nvidia4 months ago
- AMD's New Sense of Urgency: MI450X, Chance to Beat NVIDIA, and NVIDIA's New Moat4 months ago
- Next-Gen Ti Graphics Cards5 months ago
- Prototype of a Particularly Sustainable and Energy-Autonomous E-Bike Terminal Developed at HKA5 months ago
- Nvidia writes off $5.5 billion in GPUs as US gov't chokes off supply of H20s to China5 months ago
- Enhancing Chitosan Films with Silanized Hexagonal Boron Nitride for Sustainable Applications5 months ago
- White Knight to save Shibaura5 months ago
- Ed Rides The Tariff Roller-Coaster5 months ago
- Pegatron NVIDIA GB300 NVL72 and More at NVIDIA GTC 20255 months ago