Start United States USA — software Samsung will help Nvidia build custom non-x86 CPUs and XPUs in a...

Samsung will help Nvidia build custom non-x86 CPUs and XPUs in a bid to stave off competition from OpenAI, Google, AWS, Broadcom, Meta, and its other key partners

55
0
TEILEN

Custom chips under NVLink Fusion remain tied to Nvidia products exclusively
Nvidia integrates Samsung Foundry to expand NVLink Fusion for custom AI silicon
NVLink Fusion allows CPUs, GPUs, and accelerators to communicate seamlessly
Intel and Fujitsu can now build CPUs connecting directly to Nvidia GPUs
Nvidia is deepening its efforts to make itself indispensable in the AI landscape by broadening its NVLink Fusion ecosystem.
Following a recent collaboration with Intel, which allows x86 CPUs to connect directly to Nvidia platforms, the company has now enlisted Samsung Foundry to help design and manufacture custom CPUs and XPUs.
The move, announced during the 2025 Open Compute Project (OCP) Global Summit in San Jose, shows Nvidia’s ambition to extend its control across the full hardware stack of AI computing.Integrating new players into NVLink Fusion
Ian Buck, Nvidia’s Vice President of HPC and Hyperscale, explained NVLink Fusion is an IP and chiplet solution designed to integrate CPUs, GPUs, and accelerators seamlessly into the MGX and OCP infrastructure.
It enables direct, high-speed communication between processors within rack-scale systems, aiming to remove the traditional performance bottlenecks between computing components.
During the summit, Nvidia revealed several ecosystem partners, including Intel and Fujitsu, both now capable of building CPUs that communicate directly with Nvidia GPUs over NVLink Fusion.

Continue reading...