Thought Leadership
Nvidia has announced that Lenovo and Acer will launch the first Windows PCs featuring its proprietary RTX Spark chip in October.
With this, the chipmaker is pushing ahead with its effort to bring AI capabilities directly into laptops and desktop computers rather than serving them exclusively via the cloud. Tech corporations are increasingly relying on locally executed AI to reduce the costs of cloud-based services while simultaneously improving data privacy and response speed, as tasks are processed directly on users’ devices. To achieve this, the RTX Spark chip combines a Blackwell graphics processor with a Grace central processor and was developed in collaboration with the Taiwanese company MediaTek.
The full specification includes 6,144 Blackwell RTX GPU cores as well as 20 Grace CPU cores and is designed to deliver up to 1 petaflop of AI computing power with up to 128 gigabytes of shared memory, complemented by a lower-tier variant intended exclusively for laptops with 5,120 GPU and 18 CPU cores. During the presentation at IFA 2026, Acer already showcased a compact desktop concept, while Lenovo announced the Yoga Pro 9n as well as the Yoga 9n as a 2-in-1 device. Overall, devices from more than half a dozen manufacturers are expected to launch in October.
Nvidia had already unveiled RTX Spark in June as part of what it described as a roughly three-year collaboration with Microsoft, with the stated goal of reinventing the classic PC for the era of personal AI agents. Nvidia founder and CEO Jensen Huang described the undertaking during the original unveiling as follows:
“The PC is being reinvented.”
Jensen Huang, Nvidia Founder and CEO
Instead of launching and operating applications individually as before, the computer should in the future respond directly to spoken or written requests and perform the actual work independently.
Nvidia: Additional tool pools computing power of multiple devices
Alongside the launch announcement, Nvidia also presented a free software tool at IFA that can distribute independent AI requests across multiple compatible devices already present in the same network to pool their combined computing power for local AI processing.
(Editorial Team)
