Nvidia Introduces NVLink Fusion to Support AI Development

(MENAFN) Nvidia, a leading American tech firm, has introduced a new initiative titled the NVLink Fusion program, as announced by CEO Jensen Huang on Monday.

This strategic development will permit clients to incorporate rival firms’ chips within data centers that utilize Nvidia’s own foundational technologies.

According to a release from Nvidia, the NVLink Fusion strategy is designed to give users and collaborators the flexibility to operate non-Nvidia central processing units (CPUs) and graphics processing units (GPUs) alongside Nvidia’s hardware and its proprietary NVLink communication interface.

“NVLink Fusion is so that you can build semi-custom AI infrastructure, not just semi-custom chips,” Huang stated during a technology summit in Taiwan.

This highlights a shift from merely customizing chips to crafting flexible, AI-specific system architectures.

He further emphasized that through NVLink Fusion, it becomes possible to connect Nvidia processors with a diverse range of CPUs and application-specific integrated circuits (ASICs).

This approach broadens the potential for creating varied AI environments.

“In any case, you have the benefit of using the NVLink infrastructure and the NVLink ecosystem,” Huang added, underlining the continued advantage of utilizing Nvidia’s established network and tools.

In addition to the program, Huang disclosed that Nvidia is set to open a fresh branch in Taiwan.

This new office will work jointly with Foxconn, recognized as the globe’s largest electronics producer, to develop an advanced AI supercomputing system.

“We are delighted to partner with Foxconn and Taiwan to help build Taiwan’s AI infrastructure and to support TSMC and other leading companies to advance innovation in the age of AI and robotics,” he remarked, reflecting Nvidia’s broader regional investment.

Furthermore, Nvidia shared on Sunday that several major tech collaborators are already engaged in the NVLink Fusion project.

These partners include MediaTek, Marvell, Alchip, Astera Labs, Synopsys, and Cadence, all contributing to the development of cutting-edge AI chips compatible with the initiative.

MENAFN20052025000045017167ID1109569980