Nvidia has two new N1X chips coming with up to 128GB of unified memory and serious AI compute power

Digital Trends reports that NVIDIA has two N1X processors planned for its RTX Spark family, expanding the platform beyond a single flagship configuration. The chips combine an Arm-based Grace CPU with Blackwell graphics and unified LPDDR5X memory, allowing the processor and GPU to draw from the same pool. The higher-capacity designs are expected to support as much as 128GB of unified memory, a headline aimed at local AI users who need to keep larger models on the device.

The report presents the two-chip approach as a way to give RTX Spark system builders more flexibility across laptops and compact desktops. NVIDIA's public platform claims include substantial CUDA graphics resources and high AI throughput, but those figures do not describe the performance of every N1X configuration. Power limits, cooling, memory tier, firmware, production drivers, Windows on Arm compatibility, and application support will determine how the chips behave in finished products. The article's specifications should therefore be treated as an announcement and roadmap signal until retail hardware is independently tested.

For Canadian buyers, a second N1X option could broaden the range of RTX Spark systems beyond the most expensive 128GB machines, but the report does not establish Canadian pricing, retailer availability, or launch timing for each configuration. Developers and creators should compare the exact CPU and GPU configuration, usable shared memory, storage, battery capacity, and CUDA support when manufacturers publish local listings. Independent reviews will be especially important for separating NVIDIA's AI-compute claims from sustained real-world performance.