NVIDIA Officially Enters PC Market: RTX Spark Unveiled At Computex 2026

Computex 2026 and GTC Taipei will go down in history as the moment NVIDIA used to officially announce its entrance into the PC market. During his keynote at the Taipei Music Center, CEO Jensen Huang announced the RTX Spark – formerly codenamed N1 and N1X – which will power an array of premium laptops and small form factor systems coming this fall from a multitude of partners. Desktops and workstations running Windows, powered by the GB300, are coming as well.

For nearly a year, we’ve been discussing rumors of NVIDIA’s N1 and N1X. They’re fundamentally similar to the GB10 SoC powering the existing DGX Spark mini-AI workstation, but come with different configurations and are tuned specifically for lower-power mobile form factors.

The NVIDIA RTX Spark Will Power New Windows And Linux PCs

The NVIDIA RTX Spark will target an array of price and performance levels to power a diverse assortment of system types. Like the company's discrete GPUs, the various RTX Spark configurations will work with the same software, drivers, utilities, and toolsets--the lower-end configurations will just cost less and offer lower performance.

A full-featured RTX Spark will be similar to the GB10 powering the DGX Spark. The chip features 20 CPU cores (comprising 10× Arm Cortex-X925 performance cores and 10× Arm Cortex-A725 efficiency cores), with clock rates peaking at 4.1GHz. The CPU complex is connected to a Blackwell-architecture GPU with 6,144 shader cores, which is right in-line with a discrete GeForce RTX 5070. It is a two-chiplet SoC, with one Blackwell GPU chiplet paired to a Mediatek-designed CPU + I/O chiplet, linked up to each other by a very fast silicon bridge interconnect capable of 600 GB/sec of bandwidth, faster than even the latest PCI Express standards.
The chips will also use a unified memory architecture in which the CPU and GPU share the total available pool of system memory. As such, some configurations will offer much more memory capacity to the GPU, which will in turn allow the RTX Spark to handle larger AI models and potentially help in some high-resolution gaming scenarios.

Jensen highlighted the top-end RTX Spark configuration, but rumors suggest there will possibly be an 18-core (9+9) CPU version with a 5,120 CUDA core GPU. Both versions will have similar TDPs in the 45W-80W range, but you can bet the RTX Spark will boost to much higher transient power levels; a leak that hit the web a while back suggested a 245W power supply would be included with one of the higher-end NVIDIA-powered laptops coming down the pipe. Top-end RTX Spark configurations will support up to 128GB of LPDDR5X unified memory connected via a 16-channel (256-bit) interface. All RTX Spark SoCs will also offer PCI Express Gen 5 + Gen 4 connectivity as well, with various lane configurations to support SSDs, networking hardware, and other peripheral devices.


We've heard rumblings that lower power N1-based RTX Spark models will feature 12 or 10-core CPU configurations (8P+4E or 7P+3E), with 2,560 or 2,048 shader cores, respectively. Memory support on N1 configurations will top out at 64GB, using an 8-channel (128-bit) interface, and their TDPs will fall in the 45W range (but with higher boost power). These chips are likely the 400 petaflop Grace Blackwell Sparks shown on the roadmap slide below.

NVIDIA says that it is working with an array of partners to bring systems to market, including Dell, ASUS, Lenovo, HP, Microsoft, and others. No pricing was announced, but we expect the RTX Spark to arrive in more premium, creator, and AI-focused systems, with price points at the higher end of spectrum. Systems are due to arrive this fall.

Multiple Partners Will Be Offering RTX Spark-Powered Systems

NVIDIA re-enters the PC market at time when Qualcomm and Microsoft have been working to optimize Windows on Arm. We say re-enters because NVIDIA powered the original Microsoft Surface RT back in the day, but that’s a discussion for a different time. If you’ve read any of our Snapdragon X or Snapdragon X2 coverage the last couple of years, you’ll know that the Windows software ecosystem and Windows 11 on Arm have come a long way since the OS’ initial release, but there are still some compatibility issues with kernel mode drivers, and particularly with games that rely on strict anti-cheat software. This is because kernel mode drivers cannot be emulated on Windows on Arm, so any drivers or software reliant on those drivers must be compiled natively for Arm64 architectures.


Considering NVIDIA’s gaming and graphics roots, we assume the company has worked with its gaming and software partners to ensure solid compatibility and performance. NVIDIA did announce a collaboration with Adobe on RTX Spark-optimized versions of Photoshop and Premiere, with impressive AI-enhanced capabilities, and showed a handful of games running on the platform, but other than displaying a wide array of partner logos, there weren’t many software related announcements.

DLSS 4.5 Ray Reconstruction Is Coming Too

Jensen further used his keynote to announce DLSS 4.5 Ray Reconstruction. DLSS Ray Reconstruction is a part of NVIDIA’s DLSS suite that reconstructs full‑resolution ray‑traced effects (like reflections, global illumination, ambient occlusion, and so on) from lower-resolution, noisy ray samples. DLSS Ray Reconstruction combines denoising and upscaling into a single AI pass rather than relying on multiple denoising techniques. It's an alternative to traditional ray‑tracing denoisers that typically yields cleaner, more accurate ray-traced scenes, with lower compute and memory requirements.


To that end, NVIDIA announced DLSS 4.5 Ray Reconstruction, which uses a more refined transformer model trained on a larger dataset that ultimately offers improved visual fidelity with a similar impact on performance as previous versions of the technology. DLSS 4.5 Ray Reconstruction should be available in the August timeframe. DLSS ray reconstruction is also coming to Blender 5.3, and NVIDIA announced up to 4x frame-generation was coming to video with ComfyUI with the RTX Spark.

Computex 2026 marks a significant moment for NVIDIA. The largest company in the world (in terms of market cap), who is also the dominant player in AI infrastructure as well as discrete gaming GPUs, is entering a new market. That's objectively a big deal. It doesn’t, however, guarantee success. NVIDIA is entering the market at a horrible time due to the on-going DRAM and NAND shortage. The RTX Spark also faces some headwinds due to the current state of Windows on Arm as well as the incumbency and prevalence of x86. RTX Spark, at least at this early stage, is also targeting only portion of the total market. Still, having worked with NVIDIA since its inception, and witnessing its aggressiveness and execution, I know how formidable the company truly is. I'm sure this moment will have a significant impact on the PC market and it's going to be interesting to see how the pieces fall.