Nvidia launches new chip connectivity technology to accelerate AI development.

On May 20th, Nvidia announced it will sell a new technology that helps link chips together to speed up data exchange between chips – a key element in building and deploying Artificial Intelligence (AI) tools.

At the Taipei Music Center in Taiwan (China), where Computex AI was held from May 20-23, Nvidia's Chairman and CEO, Jensen Huang, introduced a new version of his NVLink connectivity technology, called NVLink Fusion. This is a product Nvidia will provide to other chip designers to support the construction of powerful, customized AI systems with multiple linked chips.

According to Nvidia, Marvell Technology and MediaTek will use NVLink Fusion for their custom chip projects. Other partners include Alchip, Fujitsu, and Qualcomm.

However, Nvidia's business prospects in China are being "cast a shadow." In a recent interview, Huang said the company had "missed out on $15 billion in revenue" in China after the US government restricted exports of its H20 chip to the market.

In addition to launching new technologies, Mr. Huang also announced plans to build Nvidia's new headquarters in the northern suburbs of Taipei. His speech at Computex highlighted Nvidia's history of developing AI chips, systems, and software.

He stated that previously, presentations often focused on graphics chips, but now Nvidia has moved far beyond its origins as a manufacturer of graphics chips for gaming to become a leading name in the field of AI chips – a sector that has boomed since the introduction of ChatGPT in 2022.

Nvidia is also developing central processing units (CPUs) that run Microsoft's Windows operating system, using technology from Arm Holdings.

At the annual developer conference in March 2025, Huang shared Nvidia's direction to meet the changing demands of computing – from building large-scale AI models to running AI-based applications. At that time, the company announced several new generations of AI chips, including the Blackwell Ultra, expected to launch later this year.

Following the Rubin line, Nvidia plans to launch the Feynman processor in 2028. The company also launched a desktop AI chip, the DGX Spark, aimed at AI researchers. Huang stated that this line of chips is now in full production and will be ready in "a few weeks."

Computex 2025 is expected to bring together around 1,400 participating companies and will be the first major event for the Asian computer and semiconductor industry since US President Donald Trump threatened to impose tariffs to encourage companies to increase production in the US.

(VNA/Vietnam+)

Source: https://www.vietnamplus.vn/nvidia-ra-mat-cong-nghe-ket-noi-chip-moi-de-thuc-day-phat-trien-ai-post1039599.vnp