San Francisco/Taipei: Qualcomm announced on Monday its plans to design custom data center central processing units (CPUs) that will integrate Nvidia’s technology to enable seamless connectivity with Nvidia’s artificial intelligence (AI) chips.
Nvidia currently dominates the AI chip market, primarily through its powerful graphics processing units (GPUs). However, GPUs are typically paired with CPUs, a sector historically led by Intel and Advanced Micro Devices (AMD). Recently, Nvidia entered the CPU market with its own offering, the “Grace” CPU, developed using Arm Holdings technology.
With this new initiative, Qualcomm is making a return to the data center CPU arena. The company initially ventured into developing Arm-based CPUs during the 2010s, including testing prototypes with Meta Platforms. However, these efforts were scaled back due to cost-cutting measures and legal challenges.
Following Qualcomm’s acquisition of a team of former Apple chip designers in 2021, the company has quietly revived its CPU development programs. Qualcomm has resumed talks with Meta about advancing data center CPU projects. Additionally, last week Qualcomm confirmed it signed a letter of understanding with Saudi Arabian AI firm Humain to collaborate on a custom data center CPU.
Also Read | New Sunni-Led Middle East Order Emerges During Trump’s Regional Visit
On Monday, Qualcomm highlighted that its forthcoming chips will incorporate Nvidia’s technology to facilitate rapid communication with Nvidia’s GPUs, which are central to its AI product lineup.
Also Read | Pro-European Nicușor Dan Clinches Victory Over Trump-Aligned Rival in Romania
“With the ability to connect our custom processors to Nvidia’s rack-scale architecture, we’re advancing a shared vision of high-performance energy-efficient computing to the data center,” said Qualcomm CEO Cristiano Amon.
