NVIDIA and Microsoft have launched the RTX Spark platform, bringing slim Windows laptops and compact desktops powered by Blackwell graphics and Arm-based Grace processors to market starting at $2,599. Available globally starting October 16, 2026, the systems target developers, creators, and early adopters with up to 128GB of unified memory and local AI agent capabilities.
The newly available NVIDIA RTX Spark laptops mark a significant shift for Windows hardware, combining NVIDIA’s full-stack AI platform, CUDA architecture, and Blackwell RTX graphics into slim, premium portable computers. Major manufacturers including ASUS, Dell, HP, Lenovo, Microsoft, and MSI are rolling out the initial wave of devices globally through their web stores and major retailers.
Hardware Configurations, Pricing, and Regional Availability
Prices for the new RTX Spark machines start at $2,599 for entry configurations, scaling up significantly for high-capacity builds. For instance, the HP OmniBook Ultra 16 starts at $3,199.99 for a 32GB RAM and 1TB SSD configuration available through Best Buy and HP.com. Stepping up to a 64GB and 2TB build lists for $4,999.99, while the top-tier 128GB and 2TB configuration requires ordering through HP.com’s configurator at $6,229.99.
In Canada, ASUS launched the ProArt P16 and P14 through the ASUS Store, Amazon Canada, Best Buy Canada, Memory Express, and Newegg Canada. The ASUS ProArt P14 lists at an MSRP of C$3,499, while ProArt P16 configurations scale across C$4,699, C$6,799, and C$8,999 depending on the chosen memory and processor trim.
The foundational superchip pairs a Blackwell RTX graphics processor with an Arm-based Grace CPU, providing up to 128GB of unified memory and delivering up to 1 petaflop of FP4 AI performance. Reviewers note that because the RAM is soldered directly to the motherboard, buyers must carefully select their memory capacity at purchase since post-purchase upgrades are impossible.
Local AI Performance and Software Ecosystem Integration
Creators want to run the best AI models on their own machines, with full control. ProArt RTX Spark PCs have the memory to run big image and video workflows locally, and ComfyUI makes those workflows easy to share, reuse and hand off to agents like Hermes.
Photo: StorageReview.com
Yoland Yan
Testing on the 128GB HP OmniBook Ultra 16 demonstrated strong multi-core and local inferencing capabilities, though reviewers observed middling single-core speeds and high thermal output under heavy load. In benchmark evaluations using MLPerf Client with Llama 3.1 8B Instruct, Windows ML averaged 54.3 tokens per second, while llama.cpp recorded 53.7 tokens per second. The multi-core evaluations reached 7,004 in Cinebench 2026 multi-core, which is 35 percent ahead of the Razer Blade 16, alongside an LLVM build completed in 206.6 seconds compared to 327.0 on the Blade. Performance figures across agentic scenarios showed RTX Spark systems completing data analyst tasks in 52.4 to 56.2 seconds and software engineering tasks in 45.3 to 48.0 seconds, while the Blade 16 required about half that duration in both tasks.
Target Audiences and Market Positioning
Industry analysts emphasize that these machines occupy a very specific niche. Tom Mainelli, Group Vice President of Device and Consumer Research at IDC, noted that These are outside the budget and needs of mainstream users, for now, pointing out that the systems target early adopters, AI developers, and content creators rather than casual buyers.
Photo: NVIDIA Newsroom
While vendors are not marketing the systems primarily as gaming rigs, the hardware supports 1440p gaming at modest settings and handles titles like Cyberpunk 2077 using NVIDIA’s upscaling and frame-generation technologies. Compact desktop versions of the platform are scheduled to follow in November 2026 from manufacturers including Acer, ASUS, Dell, Gigabyte, HP, Lenovo, and MSI.
Sophie is a tech innovator and acclaimed tech writer recognized by the Online News Association. She translates the fast-paced world of technology, AI, and digital trends into compelling stories for readers of all backgrounds.
