Nvidia's RTX Spark Arm Chip Debuts in Laptops From Six Makers

Nvidia's new Arm-based processor, the RTX Spark N1X, has started shipping inside laptops from six manufacturers, and a three-person review team at The Verge spent a week testing four of those machines to see how the chip performs outside Nvidia's own marketing, according to the outlet's published findings.

What Is the RTX Spark Chip Built to Do?

Nvidia designed RTX Spark as a high-performing Arm-based system-on-chip, or SoC, for laptops and small desktop computers, positioning it to compete with Apple's M-series processors, Intel's and AMD's x86 chips, and Qualcomm's Arm-based Snapdragon line, The Verge reported. The chip pairs its processing cores with Nvidia RTX Blackwell graphics and what the outlet described as enough unified memory capacity to run sizable AI models directly on the laptop rather than through a cloud connection.

"RTX Spark is designed to be a high-performing Arm-based SoC for laptops and small desktop computers, rivaling the best from Apple, Intel, AMD, and Qualcomm," The Verge reported.

Which Laptops Are Reviewers Testing First?

Six laptop makers have released machines built around RTX Spark, though The Verge's team has so far tested four specific configurations across three brands. Microsoft's Surface Laptop Ultra carries what the outlet called the flagship build, configured with what it described as a "staggering" 128GB of unified memory, and was reviewed in full separately by senior correspondent Tom Warren. Asus's ProArt P16 shows up in two memory tiers in the testing: a 64GB version handled by senior editor Sean Hollister and a 128GB version tested independently by laptop reviewer Antonio G. Di Benedetto, who also spent the week with a 64GB Lenovo Yoga model, according to the published account.

How Did The Verge Test the Machines?

Rather than rely on one reviewer's impressions of one machine, The Verge split the work across three staffers running parallel week-long tests on different hardware configurations from different manufacturers. The approach is designed to separate quirks specific to a single laptop's design — cooling, keyboard, display — from behavior that traces back to the RTX Spark platform itself. The outlet's own account frames the project as a collective effort spanning what it called "a small fortune's worth of powerful laptops" built on Nvidia's new chip, with the Surface Laptop Ultra review already published and the Asus and Lenovo impressions following from the week of shared testing.

What Has a Week of Testing Found So Far?

The Verge's own framing of the early results is contained in its reporting: RTX Spark amounts to "an exciting new challenger, but exclusively for the high-end." That assessment suggests the chip clears a performance bar worth taking seriously against Apple, Intel, AMD, and Qualcomm's existing laptop silicon, while the laptops carrying it — with unified memory configurations reaching as high as 128GB — are positioned toward premium buyers rather than mainstream price points. The emphasis on unified memory capacity across multiple tested units points to local AI workloads as a core use case Nvidia is targeting with the chip, rather than general productivity tasks alone.

What Is Still Unclear About RTX Spark Laptops?

The material reviewed here does not include specific benchmark scores, battery-life figures, or pricing for any of the six laptop lines carrying RTX Spark, and the published account does not detail results from the two laptop makers whose machines have not yet been tested by The Verge's team. Readers looking for a full performance comparison, including Tom Warren's complete review of the Surface Laptop Ultra and further findings from Hollister and Di Benedetto's ongoing testing, should expect those details to accumulate as the outlet continues its coverage of the platform.

Glossary

• SoC (system-on-chip): A single chip that combines a processor, graphics, and other core components, rather than housing them as separate parts on a circuit board.
• Arm-based: Describes chips built on Arm Holdings' instruction-set architecture, the same family used in Apple's M-series and Qualcomm's Snapdragon chips, as distinct from the x86 architecture used by Intel and AMD.
• Unified memory: A memory pool shared between a chip's processing and graphics cores rather than kept as separate dedicated pools, an approach that can help a device run large AI models locally.