Nvidia CEO Jensen Huang announced the company’s entry into the personal computer processor market on Monday at the Computex conference, unveiling a new Arm-based chip that will be featured in upcoming laptops from manufacturers including Microsoft, Dell, and HP. The move positions Nvidia in a market with established companies like Intel, AMD, and Apple.
During the keynote address, Huang introduced the RTX Spark superchip, which integrates the new N1X central processing unit (CPU). The chip is set to debut in the fall in a new line of Windows PCs from Microsoft, Dell, HP, ASUS, Lenovo, and MSI.
“Microsoft and Nvidia are going to reinvent the PC,” Huang said during the presentation. He added that the new computers would feature agentic AI capabilities.
The RTX Spark combines one of Nvidia’s Blackwell graphics processing units (GPU) with the custom N1X CPU, which is based on Arm architecture and was designed with Taiwanese firm MediaTek. The superchip also includes 128 gigabytes of unified memory. According to the announcement, Taiwan Semiconductor Manufacturing Company (TSMC) will produce the processor using its 3-nanometer technology.
An Nvidia spokesperson confirmed the company’s initial rollout plans include more than 30 laptop models and 10 desktop models featuring the new chip.
The development of an Arm-based PC chip by Nvidia was previously reported by Reuters in 2023. A company spokesperson said Nvidia has been working on the chip with Microsoft for “many, many years,” and described it as having higher performance and efficiency than traditional x86 processors.
The first laptops with the RTX Spark superchip will be as thin as 14 millimeters and will be positioned at a premium price point, according to the company. Nvidia stated the initial target audience includes “creators, AI developers and gamers.” Regarding performance, an Nvidia spokesperson said the chip is “roughly equivalent” to its RTX 5070 laptop GPU, with more detailed metrics to be released closer to the fall launch.
In a separate announcement at the conference, Huang stated that Nvidia’s Vera CPU for data centers is now in full production and will be available starting in the fall. He identified the Vera CPU as a “new major growth driver” for the company. Early customers for the data center chip include Anthropic, OpenAI, SpaceX’s xAI, Dell, Oracle, and CoreWeave.
Ian Buck, Nvidia’s vice president of hyperscale and high performance computing, commented on the Vera CPU’s capabilities. Buck said that Vera can produce tokens 1.8 times faster than current x86 processors. “Fast CPUs have become essential to keeping the AI factory moving,” he stated.
