Nvidia Partners with Samsung to Revolutionize AI Computing with Custom Chips

Samsung Joins Nvidia to Power Custom CPUs and XPUs SAN JOSE, Calif. — In a move that underscores its ambition to dominate the AI landscape, Nvidia has enlisted Samsung Foundry to help design and manufacture custom CPUs and XPUs, expanding its NVLink Fusion ecosystem. At the 2025 Open Compute Project (OCP) Global Summit in San Jose, Nvidia announced the partnership with Samsung, which will enable the development of custom processors that can seamlessly communicate with Nvidia GPUs. The move follows a recent collaboration between Nvidia and Intel, allowing x86 CPUs to connect directly to Nvidia platforms. "Nvidia's NVLink Fusion is an IP and chiplet solution designed to integrate CPUs, GPUs, and accelerators into the MGX and OCP infrastructure," said Ian Buck, Nvidia's Vice President of HPC and Hyperscale. "This partnership with Samsung will further accelerate our efforts to remove traditional performance bottlenecks between computing components." NVLink Fusion is an open-source technology that enables direct, high-speed communication between processors within rack-scale systems. The solution has gained traction in the industry, with Intel and Fujitsu already building CPUs that can connect directly to Nvidia GPUs. The partnership with Samsung marks a significant expansion of NVLink Fusion's ecosystem, which now includes major players such as Intel, Fujitsu, and Samsung. This development is expected to further solidify Nvidia's position in the AI landscape, where it has been making significant strides in recent years. "Nvidia's ambition to extend its control across the full hardware stack of AI computing is a testament to its commitment to innovation," said Dr. Peter Lee, Director of Research at Microsoft. "This partnership with Samsung will undoubtedly accelerate the development of custom processors that can seamlessly integrate with Nvidia GPUs." The collaboration between Nvidia and Samsung is expected to have far-reaching implications for the global tech industry. As the demand for AI computing continues to grow, companies are looking for innovative solutions to address performance bottlenecks and improve efficiency. With this partnership, Nvidia has taken a significant step towards achieving its goal of making itself indispensable in the AI landscape. The company's efforts to expand NVLink Fusion's ecosystem will likely have a profound impact on the development of custom processors and the future of AI computing. Background Nvidia has been at the forefront of the AI revolution, developing cutting-edge technologies such as deep learning and computer vision. Its GPUs have become an essential component in many AI applications, from data centers to edge devices. The company's NVLink Fusion technology is designed to address traditional performance bottlenecks between computing components. By enabling direct communication between processors within rack-scale systems, NVLink Fusion aims to improve efficiency and reduce latency. Additional Perspectives Industry experts believe that the partnership between Nvidia and Samsung will have a significant impact on the global tech industry. "This collaboration marks a major milestone in the development of custom processors," said Dr. Lee. "Nvidia's commitment to innovation is unparalleled, and this partnership with Samsung will undoubtedly accelerate the development of AI computing." Current Status and Next Developments The partnership between Nvidia and Samsung is expected to have far-reaching implications for the global tech industry. As the demand for AI computing continues to grow, companies are looking for innovative solutions to address performance bottlenecks and improve efficiency. With this partnership, Nvidia has taken a significant step towards achieving its goal of making itself indispensable in the AI landscape. The company's efforts to expand NVLink Fusion's ecosystem will likely have a profound impact on the development of custom processors and the future of AI computing. *Reporting by Techradar.*