Nvidia CEO Jensen Huang unveiled the RTX Spark laptop during his keynote speech at Computex 2026 in Taipei on June 1, 2026.
The hwa Cheng | Afp | Getty Images
Nvidia has emerged as the world’s most important company by dominating the market for AI chips in the data center. Now the company is expanding its power to chips that will serve as the main processor of personal computers, entering a field that has long been dominated by Intel, Advanced Micro Devices, Qualcomm again an apple.
During a keynote speech at Taiwan’s Computex conference on Monday, Nvidia CEO Jensen Huang introduced the new N1X processor made on the side. Microsoft. It will be included in the new RTX Spark superchip, which debuts this fall in the new line of Windows PCs from Microsoft, Dell, HP, ASUS, Lenovo and MSI.
“This renaissance of the computer is as big a story as the renaissance of the phone in what we now know as a smartphone,” Huang said, pointing to the fact that AI will be running on all new computers.
“Microsoft and Nvidia will reinvent the PC,” he added. “This is the first completely redesigned, reinvented line of PCs in 40 years.”
Nvidia’s initial plan is to gradually release more than 30 laptops and 10 desktops with the new chip over time, an Nvidia spokesperson said.
The first PC processor is made of two Flagship types of Nvidia chips combined together, and 128 gigabytes of integrated memory. It pairs one of Nvidia’s Blackwell graphics processing units with a new Arm-based N1X central processing unit, custom designed by Taiwanese company MediaTek.
RTX Spark represents a major shake-up in the PC industry, which is already undergoing significant changes driven by the AI boom. Arm-based processors like Nvidia’s are gaining power over traditional x86 processors backed by Intel and AMD, while the overall market for CPUs is exploding into what Huang says will be a $200 billion industry.
This reinvention of the computer is as big a story as the reinvention of the phone to what we now know as a smartphone.
Nvidia told CNBC in February that CPUs were “becoming a bottleneck” during AI workflows. The following month, Nvidia unveiled an entire rack full of Vera CPUs for data centers. While training large models requires a large number of parallel calculations – a task best for a GPU – accessing that data and pushing it to multiple agents requires a standard CPU-powered computer.
Nvidia’s new PC processor will be used Taiwan Semiconductor Manufacturing Company3-nanometer technology, currently available only in Taiwan.
Anticipation for Nvidia’s Arm-based PC has been building for years. Reuters reported that the company was working on a PC chip by 2023, as part of the push Microsoft getting companies to make Arm-based processors for their computers. An Nvidia spokesman said it had been working on the chip with Microsoft for “many, many years,” adding that it would be “far, far more capable, higher performance, more efficient” than traditional x86 processors.
Intel is an early pioneer of the x86 instruction set, starting in the 1970s. Intel unveiled its new data center Xeon 6+ CPUs at Computex in Taiwan on Monday.
Recently, a number of companies have been switching to one of Arm’s efficient architectures, which first became widespread in the original iPhone in 2007.
Apple is now making Arm-based processors for its computers, launching a line of MacBooks with its latest M5 chips in March. Arm also unveiled its first in-house CPU that month, and AMD is reportedly working on an Arm-based PC.
The first laptops powered by Nvidia’s new chip will be as thin as 14 millimeters, carry a premium price tag, and will be released in other small desktop models. While the RTX Spark will eventually expand into different price points, Nvidia said it’s currently aimed at creators, AI developers and gamers, “looking for very small and light laptops, small laptops, portable laptops, or compact computers.”
Nvidia said it will release more performance metrics closer to when the chip hits the market in the fall. At the moment, the RTX Spark is “almost on par” with Nvidia’s flagship RTX 5070 GPU, according to its spokesperson.
Huang also announced at Computex on Monday that Nvidia’s Vera CPU for data centers is now in full production. Huang said Nvidia is making millions of CPUs “in a market that has never existed before.” Vera will be available from autumn. Early customers include Anthropic, OpenAI, XAI for SpaceX, Dell, Oracle and CoreWeave.
“This will be our new growth driver,” Huang said. “These CPUs will work both, but they also have to be very powerful, so that we can integrate as many CPUs as possible in the industry without taking away power from token generation.”
“Fast CPUs are critical to keeping the AI industry moving,” said Ian Buck, Nvidia’s VP of hyperscale and high performance computing.
Buck said Vera can generate tokens 1.8 times faster than x86 today, “improving overall agent token performance, enabling smarter, longer-term agents and ultimately, generating more data center token revenue.”
WATCH: An exclusive preview of Nvidia’s Vera Rubin
