Qualcomm has announced plans to utilize NVIDIA's technology to produce custom data center central processing units (CPUs) designed to connect with NVIDIA's dominant artificial intelligence (AI) chips. While NVIDIA's chips lead the AI market, they are typically paired with CPUs, a market traditionally led by Intel and Advanced Micro Devices (AMD). NVIDIA has also ventured into the CPU market by designing its own "Grace" CPU using Arm Holdings' technology.
Qualcomm revealed its intention to re-enter the data center CPU market. Previously, Qualcomm developed Arm-based CPUs in collaboration with Meta Platforms but halted these efforts due to cost and legal challenges. However, after acquiring a team of former Apple chip designers in 2021, Qualcomm quietly resumed its CPU development and re-engaged discussions with Meta regarding data center CPUs.
Qualcomm confirmed last week that it signed a memorandum of understanding with Saudi Arabian AI company Humain to develop custom data center CPUs. Qualcomm's future chips will incorporate NVIDIA's technology to facilitate rapid communication with NVIDIA's graphics processing units (GPUs), which are integral to Qualcomm's AI chip portfolio. Qualcomm CEO Cristiano Amon emphasized the collaboration's aim to advance high-performance, energy-efficient computing in data centers.
