Nvidia debuted its RTX Spark AI superchip at Computex 2026 in Taipei on June 1, aiming to enable local AI capabilities on Windows PCs and laptops [1, 2, 3, 4, 5, 6, 7]. The chip integrates a 20-core ARM CPU, up to 6,144 Blackwell GPU cores, and supports as much as 128GB of unified LPDDR5X memory [8, 4, 6, 7].

RTX Spark laptops and compact desktops will launch from major PC makers including Dell, Lenovo, Asus, HP, Microsoft, MSI, Acer, and Gigabyte starting in fall 2026 [1, 2, 8, 4, 5, 6, 7]. The chip was co-developed with Taiwan’s MediaTek and is designed to replace traditional mouse and keyboard inputs by running autonomous AI agents locally on the PC [1, 2, 5].

Nvidia CEO Jensen Huang called RTX Spark a reinvention of the PC after four decades. He said, "The PC is being reinvented. For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask — and the PC does the work" [3]. Huang also noted, "Microsoft and Nvidia meticulously optimised everything" for the platform [2]. Windows 11 and the ARM version of Windows will be specifically optimized to support RTX Spark’s AI agent features [1, 2, 4, 6].

RTX Spark supports Nvidia RTX gaming technologies such as ray tracing, DLSS, and Reflex. Some sources report it can run games at 1440p resolution and over 100 frames per second [9, 3, 7]. However, analysts caution ARM-based chips may face gaming compatibility challenges on Windows due to app and anti-cheat software issues [4, 7].

Industry experts are optimistic about RTX Spark’s potential. Neil Shah from Counterpoint Research said the chip aims to transform traditional PCs into "real useful agentic AI personal computers that will eventually be in every home" [1]. Bloomberg analyst Steven Tseng noted that Spark "raises the bar by arguing that a true AI PC should be able to run AI agents locally," in contrast to PCs relying on cloud AI [2].

Nvidia’s RTX Spark marks the company’s reentry into the consumer PC processor space, taking on Intel and AMD with a powerful ARM-based chip [1, 2, 5]. It runs at one petaflop of AI compute power, reinforcing Nvidia’s broader AI PC ecosystem which also includes the Vera CPU used by OpenAI, Anthropic, and SpaceX [1, 3, 7].

Nvidia is already developing follow-up chips named N2X and N3X, along with a smaller N1 variant, to expand the platform [10]. The first RTX Spark-based devices are expected to reach consumers in fall 2026.