By 
Nvidia is planning to sell its NVLink Fusion technology to other companies, enabling faster communication between artificial intelligence chips. This technology allows chip designers to build powerful custom AI systems by linking multiple chips together, which is essential for developing and deploying AI tools.
Nvidia’s CEO, Jensen Huang, announced the plan at the Computex AI exhibition in Taipei, where he discussed the company’s history of building AI chips and systems. Huang noted that Nvidia has shifted its focus from graphics chips to AI chips, driven by the demand sparked by applications like ChatGPT.
The NVLink Fusion technology will be adopted by companies like Marvell Technology and MediaTek for their custom chip projects. Nvidia developed NVLink years ago, and it’s used to exchange large amounts of data between chips, such as in the company’s GB200 model.
By selling NVLink Fusion, Nvidia is positioning itself as a key player in the AI hardware ecosystem. This move could help the company generate new revenue streams and solidify its dominance in the AI chip market. Nvidia is also designing CPUs compatible with Microsoft Windows using Arm Holdings architecture.
Nvidia’s upcoming products include the Blackwell Ultra chip, expected later this year, followed by the Rubin series and Feynman processors in 2028. The company has also launched a desktop version of its AI chips, called DGX Spark, targeting AI researchers. DGX Spark is already in production and will be ready in a few weeks.
Nvidia’s goal is to stay at the center of the AI boom by providing the technology and infrastructure needed for fast and efficient AI processing. With its expanding chip roadmap and growing role in the AI ecosystem, Nvidia continues to shape the future of computing.
Comments