NVIDIA’s Grace CPU C1 Takes Center Stage at COMPUTEX

The AI Powerhouse Redefining Efficiency and Performance

NVIDIA is making waves at COMPUTEX in Taipei with its Grace CPU C1, a chip designed to push the boundaries of AI and high-performance computing. Backed by manufacturing giants like Foxconn, Jabil, and Supermicro, the Grace lineup—including the Grace Hopper Superchip and Grace Blackwell—is poised to dominate the next wave of AI infrastructure. With cloud providers already adopting the monstrous Grace Blackwell NVL72 (which packs 36 Grace CPUs and 72 Blackwell GPUs), NVIDIA’s vision for AI acceleration is clearer than ever.

“Grace isn’t just another CPU—it’s a leap forward in energy efficiency and raw compute power,” says an NVIDIA spokesperson.

The Grace CPU C1 isn’t just about brute force; it’s about doing more with less. Delivering a 2x energy efficiency boost over traditional CPUs, it’s tailor-made for edge computing, telco, storage, and cloud deployments. This isn’t just theoretical—partners like WEKA and Supermicro are already integrating Grace into storage solutions, leveraging its high memory bandwidth to tackle data-intensive AI workloads. Meanwhile, the NVIDIA Compact Aerial RAN Computer combines the Grace CPU C1 with an L4 GPU and ConnectX-7 SmartNIC, bringing AI-RAN capabilities to telecommunications.

At NVIDIA GTC Taipei, running May 21-22 alongside COMPUTEX, the company will showcase these breakthroughs in detail. From AI training to inference, Grace is proving to be the backbone of tomorrow’s infrastructure. With major manufacturers lining up to build systems around it, NVIDIA’s bet on Grace isn’t just a gamble—it’s a blueprint for the future.