Newsletter Subscribe
Enter your email address below and subscribe to our newsletter
Enter your email address below and subscribe to our newsletter

Nvidia officially unveiled the RTX Spark at its GTC Taipei keynote on Sunday evening, marking the company’s first consumer Windows PC processor in over a decade and a direct challenge to Intel, AMD, and Qualcomm in the laptop market.
CEO Jensen Huang announced the chip during his keynote at the Taipei Music Center, calling it “the most efficient PC chip ever built.” The RTX Spark, based on Arm architecture and co-developed with MediaTek, features a 20-core custom CPU paired with 6,144 CUDA cores built on Nvidia’s Blackwell graphics architecture — GPU performance roughly equivalent to an RTX 5070 discrete graphics card. The chip supports up to 128GB of unified LPDDR5X memory shared between the CPU and GPU, enabling users to run large AI models locally.techradar
The chip is derived from the same GB10 Superchip architecture that powers Nvidia’s DGX Spark desktop AI workstation, which shipped last year, but has been redesigned for Windows laptops and desktops at lower power envelopes.tomshardware
Nvidia has secured a broad coalition of PC manufacturers for the RTX Spark launch. Dell confirmed an XPS laptop with the chip, while Lenovo, Asus, and MSI have teased or confirmed devices. Microsoft is also part of the partnership, with reports indicating new Surface hardware built around the chip will be announced at the company’s Build conference on June 2.tbreak
Lenovo’s Legion 7 gaming laptop variant will ship with a 245W power adapter, suggesting the N1X configuration is aimed at high-performance users, while an N1 variant targets mainstream efficiency laptops. The first devices are expected to reach consumers in late 2026.techflowdaily
The move places Nvidia in direct competition with Qualcomm’s Snapdragon X series, which pioneered the recent wave of Windows on Arm devices, as well as traditional x86 incumbents Intel and AMD. Nvidia’s advantage lies in its CUDA software ecosystem, which dominates AI development tools and creative applications.wccftech
The chip’s large unified memory pool positions it for running large language models with over 100 billion parameters on a laptop — a capability that could distinguish RTX Spark devices in a market where local AI processing is becoming a key differentiator. Pricing has not been officially disclosed, but analysts expect premium positioning above $3,000 for top configurations with 128GB of memory.wccftech