Nvidia enters Arm-based PC market with RTX Spark chip targeting AI workloads and gaming

Nvidia gets into the Arm PC business with new high-end RTX Spark processor

These days, Nvidia primarily sells AI data center products, and its traditional consumer devices feel like more of a side project. But the company occasionally still releases something designed for consumers. After a couple of years of rumors, Nvidia has announced an Arm-based chip designed to power Windows PCs. Dubbed RTX Spark, the new chip combines a 20-core Nvidia Grace CPU co-developed with MediaTek, up to 6,144 Blackwell-based GPU cores (the same architecture as the RTX 50-series GPUs), and support for up to 128GB of unified LPDDR5x memory. Nvidia and its partners offered nothing about expected pricing, but both "slim Windows laptops with all-day battery life and premium displays" and "compact desktop PCs" are slated to be "available this fall" from partners including Asus, Dell, HP, Lenovo, Microsoft, MSI, Acer, and Gigabyte. This isn't Nvidia's first chip for Windows PCs; earlier chips in the Tegra series powered several of the short-lived Windows RT tablets. But Tegra chips largely stopped appearing in consumer devices following the Tegra X1 in the late 2010s (variants power the original Nintendo Switch and the apparently unkillable Nvidia Shield TV box). Modern Arm-based PCs in the Windows 10 and Windows 11 eras have all used processors from Qualcomm. Nvidia stands to benefit from the years of progress made on the Arm version of Windows since the Windows RT days. Microsoft's x86-to-Arm code translation layer, codenamed Prism, has gotten better and faster. Many major apps now ship Arm-native versions that can run without the performance and responsiveness penalties that still crop up in translated apps. Most of the time, at least for productivity work and general computing, an Arm-based PC and an Intel or AMD-based PC look and feel indistinguishable. And Nvidia's entry into the market may help improve the gaming experience, one area where the Arm version of Windows still falls short. Translated games often play, but they can show lag or responsiveness issues even when running at a reasonable frame rate; many games that require kernel-level anti-cheat software to be installed still don't run at all. Nvidia and Microsoft told The Verge that they were actively working with Riot Games to support League of Legends and Valorant on Arm PCs; with Krafton to support PUBG; and with the developers of Easy Anti-Cheat, BattlEye, and Denuvo. What we know about specs The RTX Spark appears to be a consumer rebrand for the silicon Nvidia launched late last year as the DGX Spark, the heart of a tiny developer workstation for people working with AI models. And while that desktop is about as high-specced as an RTX Spark system might get -- it includes 128GB of RAM and a 4TB SSD -- its current $4,699 price tag suggests that the fastest RTX Spark machines won't come cheap. (That's also, for the record, already $700 more than the box's $3,999 launch price, a reminder of the RAM and storage supply crunch that Nvidia has helped drive with its AI data center products.) Knowing the DGX Spark's specifications gives us a better idea of how RTX Spark will perform, at least in its most capable form. The Nvidia Grace CPU combines 10 high-performance Arm Cortex-X925 CPU cores and 10 medium-sized Cortex-A725 core; Arm makes a smaller, higher-efficiency Cortex-A520 core, but it isn't used here. That makes the RTX Spark a bit more like Apple's M5 Pro or M5 Max, which use a mix of medium-sized performance cores and large "super" cores without any of the M5's smaller efficiency cores. Having 6,144 Blackwell-based GPU cores puts the RTX Spark's GPU on the same level as the desktop version of the GeForce RTX 5070, well above the mobile version of the RTX 5070 (4,608 cores) but below the mobile version of the RTX 5080 (7,680 cores). The GPU's performance will be limited somewhat by the size of the power envelope in laptops and mini PCs (Nvidia says RTX Spark's power use maxes out at 80 W, whereas a desktop 5070 can consume up to 250 W by itself), and by using slower LPDDR5x memory instead of the GDDR7 RAM that RTX 50-series GPUs use. But for some games and for AI and machine learning workflows, the relative slowness of the RAM will be offset by the amount you can get -- unified memory means the CPU and GPU can both access almost all of the RAM available in the system, giving users access to over 100GB of VRAM instead of the paltry 8GB or 12GB that you get with an RTX 5070. Unified memory pools attached to good-enough GPUs have also made systems like Apple's Mac mini and Mac Studio and the Ryzen-based Framework Desktop useful platforms for developers and people looking to run AI models locally. The Blackwell-based architecture will also mean that RTX Spark will be able to take advantage of the same DLSS upscaling and frame generation features as other RTX 50-series GPUs, including the controversial upcoming DLSS 5 release. As for the lower-end RTX Spark models, Nvidia didn't have much to say, but we can glean some details. Rumors and leaked specs (via VideoCardz) suggested that Nvidia was working on two distinct pieces of silicon, one codenamed "N1X" and one codenamed "N1." Both chips would come in two different flavors -- one with all CPU and GPU cores fully enabled and one with some defective cores disabled, a process called "binning" that is commonly used to maximize the amount of sellable silicon you get out of a manufacturing run. The fully enabled N1X aligns with the specs for the RTX Spark; there's also allegedly an N1X version with 18 CPU cores and 5,120 CUDA cores. The slower, smaller N1 chip is said to include up to 12 CPU cores (8 Cortex-X925 and 4 Cortex-A725) and 2,560 CUDA cores, which is the same number as the desktop GeForce RTX 5050. A binned version of that chip is also said to exist, with 10 CPU cores and 2,048 CUDA cores. The basic N1 is said to support up to 64GB of RAM and have a lower maximum power draw of 45 W. This would put its power envelope closer to something like Intel's Core Ultra Series 3 (codenamed Panther Lake) chips, implying we could see versions of RTX Spark appear in premium thin-and-light laptops eventually.

Nvidia RTX Spark May Light a Fire for Windows on Arm

Buckle up: Nvidia is "reinventing the personal computer," according to CEO Jensen Huang. Microsoft and Nvidia have been cozying up to one another in preparation for Nvidia's highly anticipated launch of the RTX Spark. It's a new Arm-based system-on-chip (or "SoC") platform that brings Nvidia's Blackwell architecture to thin and light Windows laptops and mini desktops. The goal is to provide high-power processing performance for running personal agents, creative work and gaming, but without the space, power needs and cooling requirements usually imposed by discrete graphics. The RTX Spark joins Qualcomm's Snapdragon X processors running Windows on Arm, with similar claims of "all-day battery life." Snapdragons achieve that, but one thing to remember about Nvidia's chip is that it's intended for far heavier workloads than Snapdragon processors. Those aren't meant to "render ultralarge 90GB-plus 3D scenes, edit 12K 4:2:2 video, generate 4K AI videos, run 120B-parameter LLMs with up to 1 million tokens context using agents locally, and play AAA games at 1440p and over 100 frames per second," all of which can tank your battery life. It remains to be seen if the Spark can live up to that under normal usage. This is the first of what Nvidia says it plans to be a line of chips across a variety of price segments. These first models are slated to ship this fall: * Microsoft Surface Laptop Ultra * Dell XPS 16 * Asus ProArt P14 and P16 * HP Omnibook X 14, Omnibook Ultra 16 * Lenovo Yoga Pro 9n * MSI Prestige N16 Flip AI The 15-inch Surface Laptop Ultra is particularly notable because Microsoft hasn't updated its screens in far too long, and the Surfaces (both desktop and laptop) never incorporated discrete GPUs their prices seemed to demand. The Ultra has a higher-resolution (262ppi) 15-inch mini LED touchscreen that supports HDR (with peak brightness of 2,000 nits), unlike the older, meh model. Microsoft hasn't updated its Surface Laptop Studio in three years, and this is the chip and screen it needs if Microsoft plans to bring it back from the dead. There will also be mini desktops. It seems to have been a resurgence of these -- at least an increase in the number of manufacturers offering them -- thanks to developers. The RTX Spark models will compete with AMD Ryzen AI Halo-based models for example. They're expected from companies such as Acer, Asus, Dell, HP and Lenovo, among others. Nvidia's planning to have a desktop, laptop and workstation for each generation of chips. Given current price volatility, we won't know how much they'll cost until they're closer to shipping. AI's ravenous demand for components -- and the resources needed to make them -- has created severe shortages of memory, processors and SSD storage, driving computer and phone prices higher and even affecting available configuration options. Spark it up The chip is an offshoot of the DGX Spark (GB10), which powers Linux-based compact desktops specifically targeted at developers and now Windows-based DGX Station. The Spark was designed in conjunction with MediaTek, and has similar specs to the DGX: 6,144 CUDA cores, a 20-core Grace CPU, ability to access up to 128GB RAM and more. Nvidia says it supports up to 120B parameter agents with a 1M context. (For reference, AMD says its top Ryzen AI Max Pro 400 series chip can can handle up to 300B parameter models). Its GPU specs are more or less comparable to an RTX 5070, but the unified memory architecture means it has access to a lot more RAM than 12GB. Nvidia says that system configurations can go as low as 16GB, though, which means it could potentially bottleneck when a dedicated 5070, with 12GB VRAM, might not. The company gave 100fps 1440p as its reference for gaming performance (though it wasn't clear whether that was with or without DLSS 4.5 enabled). Nvidia claims the chip's overall AI performance is one PFLOPS (a billion floating point operations per second), but that's based on FP4 calculations. On one hand, FP4 is the current darling of the data formats because it's faster than the other floating point choices and more accurate than integer, but there are some tradeoffs. (Procyon has a great visual example of what speed versus accuracy tradeoffs can mean for image generation.) But among the consumer SoCs, this is the first to support it in hardware. The real competition for these is the M5 Pro and M5 Max MacBook Pros, which target the same users, but the M5 line doesn't support FP4 and FP8 data types, which may turn out to be a hindrance. The part itself can run at anywhere from "single digits" to 80W, which means you'll really need to pay attention to whether a laptop runs at full power or if the manufacturer is throttling it. In other words, it sounds like performance, especially on battery, may vary a lot. Typically, mobile processor power envelopes are smaller bands; for instance, the Intel Core X9 388H specifies 15W-85W. It has an NPU, which Nvidia doesn't seem to want to talk much about, but the systems with the Spark are considered Copilot Plus-qualifying, so it must be able to hit at least 40 TOPS. RTX Spark might seem powerful, but Nvidia is maintaining its strict division between pro and consumer markets. For instance, it doesn't plan to run a certification program for applications or support ECC memory. In addition to being one of Nvidia's launch partners with its Surface Laptop Ultra, Microsoft has been working to make the necessary updates to Windows in order to take advantage of the new chip. Like Qualcomm's Snapdragon X series processors, Windows doesn't natively support the Arm instruction set the way it does Intel and AMD's x86-architecture chips, which were foundational to the PC. Instead, Arm-based systems use an emulation layer called Prism to translate instructions. Emulation is partly why the early systems based on Qualcomm chips experienced performance and compatibility problems. Windows modifications Many of the updates to Windows that are necessary to support the hardware are under the hood, but one will be right in your face: Microsoft's putting Spark-run agents on the Taskbar. A lot of the changes we've seen in Windows recently have been laying the groundwork for this. Prism was written specifically for Qualcomm's SoCs, since it was the only Arm-based silicon the operating system needed to run on. Supporting the RTX Spark meant updating Prism and other core parts of Windows to efficiently distribute workloads across the CPU cores, balance cooling and performance, address and intelligently manage a larger amount of the unified memory available to the GPU (for AI processing with TensorRT) and more. Qualcomm doesn't have nearly as much invested in Windows gaming performance as Nvidia does, for obvious reasons. For example, Nvidia has been working with Microsoft to improve compatibility with anti-cheat software (such as Epic's Easy Anti-Cheat), which has prevented some games from running on the devices, as well as support for the Xbox app, which is key to Microsoft's game-on-everything strategy. Adobe is also reengineering parts of its imaging engines to tap into the Spark directly, notably with several new pipelines to accelerate more GPU- and AI-intensive features such as rendering complex timelines in Premiere Pro and improving natural brushes in Photoshop. While CUDA and TensorRT already operate on Nvidia's discrete mobile GPUs, taking optimal advantage of them on this different architecture requires some rejiggering. The applications will also be able to interact with Windows agents. Plus, Nvidia is porting OpenShell -- its security protocols for running agents -- to Windows, via new controls that Microsoft will reveal at its Build conference in the first week of June. OpenShell, in theory, lets you define guardrails for your agents, route queries to approved local models based on your privacy policies and let it "disguise" personal information when querying cloud-based models. Nvidia is trying to expand everyday agenting beyond developers, with the notion that "broad adoption has been limited by the inability to run agents securely and privately on users' primary PCs." I suspect the trust issues are more complicated than that. The company says that OpenShell will be incorporated into the current agenting faves, OpenClaw and Hermes.

This could be Windows' M1 moment -- but expect it to cost a ton

Nvidia's announcement that it's getting into the consumer laptop chip space with RTX Spark is huge. Apple has proved for years that Arm-based chips can perform incredibly well while also delivering great battery life -- at least on the Mac. In the Windows world, performance hasn't fully matched up under Qualcomm chips, mostly in the graphics department. There's clearly still untapped potential, and Nvidia seems to be promising to deliver it. This could be Windows' moment to blow us away with a new generation of supremely capable chips, much like Apple's back in 2020, with the introduction of the M1. But why does this launch feel simultaneously exciting and fraught in 2026? The Nvidia RTX Spark sounds like a monster of a laptop chip: 20 CPU cores, 6,144 GPU CUDA cores, and 128GB of unified LPDDR5X memory. Its integrated graphics are said to be equivalent to an RTX 5070 Laptop GPU -- though Nvidia notably has shown nothing of performance metrics or actual benchmarks. As my colleague Sean Hollister pointed out, it's basically a GB10 chip from Nvidia's DGX Spark mini-PC. Nvidia calls it a "superchip" and "the most efficient PC chip ever built," while Microsoft is billing its Spark-equipped Surface Laptop Ultra as "the most powerful thing we've ever made." It should surprise no one that Nvidia CEO Jensen Huang spent most of his time introducing RTX Spark laptops talking about AI and agents. The majority of the two-hour Nvidia keynote was about agents and "CPUs for agents," which Huang said is Nvidia's "new major growth driver." But beyond the local AI compute that RTX Spark laptops will be capable of, they're also being aimed at creators. Adobe is even onboard with optimized versions of Photoshop and Premiere. This is Nvidia, Microsoft, and Windows laptop makers aiming squarely at Apple's MacBook Pros. It's not clear yet which MacBook Pro (the M5, M5 Pro, or M5 Max), but these laptops are looking like they're going to be expensive. The lineups announced so far for the fall includes the Surface Laptop Ultra, Dell XPS 16, Asus ProArt P14 and P16, Lenovo Yoga Pro 9n, MSI Prestige N16 Flip AI Plus, HP OmniBook Ultra and OmniBook X 14, and unnamed models from Acer and Gigabyte. Existing or similar models from this lineup typically start at $2,000 to $2,500 and up (aside from some more modest configs of the OmniBook X). This isn't surprising considering the RTX Spark's 128GB of RAM. If you look at AMD's Strix Halo APU with 128GB of RAM -- the closest analog to the RTX Spark but built on x86 -- you have options like Asus' ROG Flow Z13 for an MSRP of $3,300 and ProArt PX13 GoPro Edition for $3,000. That Nvidia DGX Spark desktop with the GB10 chip the RTX Spark is based on? One of those costs about $4,700. So how much do you think a Spark laptop with 128GB of memory will cost when you also add things like a keyboard, trackpad, battery, and a 15-inch Mini LED touchscreen? Nvidia said there will be RTX Spark chips with lower amounts of RAM, but thanks to RAMageddon many laptops with 16GB or 32GB of memory are getting pricier, too -- especially as new models come out. Nvidia could blow the doors off everything else in the performance department when these laptops hit in the fall, but the difference between this and Apple's M1 moment was that Apple started with the more affordable Mac Mini and MacBook Air, along with the cheapest MacBook Pro. That meant the average buyer was able to feel the benefits right away, and a lot of early sales also meant a lot of early incentive for developers to prioritize adding support for the new chips. It took nearly another year for Apple to scale things up to the M1 Pro and M1 Max with revitalized MacBook Pros. Nvidia isn't aiming for an M1 moment as much as it's trying to skip to an M1 Max or even M1 Ultra moment. And it's doing so while computers are getting increasingly costly and consumer spending power takes a nosedive. There's a reason the MacBook Neo rocked the tech world at $599. Does the same happen at $2,499? When these new laptops come out in the fall, there will be four viable chip options across a range of Windows laptops: Intel, AMD, Qualcomm, and Nvidia. It's already been nice having three choices: AMD options typically offer great performance at the cost of some battery life, Qualcomm offers the absolute best battery life and standby time but sadly poor games support, and Intel is often a balanced option that maintains full x86 compatibility. With Nvidia in the mix on Arm, we could get another option with strong battery life and much more graphics power. There's also at least a chance that gaming on Arm will grow closer to parity with the wide compatibility that x86 Windows gamers are accustomed to. Microsoft and Nvidia getting Riot Games to port their anti-cheat software to Arm for games like Valorant and League of Legends and working with other developers using Easy Anti-Cheat, BattlEye, and Denuvo is a big win for Windows on Arm. I love seeing more competition, as it's nice to have all this choice. The latest chips from Intel, AMD, and Qualcomm are each great in their own ways. I welcome an Nvidia option that performs well and has exceptional battery life but doesn't lack games like Macs do. But even if the RTX Spark ushers in sea change, the rising tide of prices is bound to leave many adrift.

Nvidia's RTX Spark Is an Earthquake for the PC Industry. Here's How I See the Landscape Shifting

Nvidia kicked Computex off with massive news: a major consumer laptop play, powering a new class of AI PCs with a laptop superchip called the RTX Spark. This isn't just about throwing an Nvidia sticker on a keyboard deck and dropping a jumped-up Arm chip inside. Nvidia's move is a step change in laptop hardware. By taking the heavy-lifting AI components of Nvidia's DGX Spark developer desktop, embracing Windows on Arm, and using the unified-memory, system-on-a-chip (SoC) playbook of the Apple MacBook Pro, this architecture dramatically levels up consumer compute. Simultaneously, it gives Microsoft the hardware foundation it needs to rewrite Windows from the ground up for deep, local agentic AI capabilities. Oh, and for good measure, it finally tackles the Achilles' heel of Windows on Arm by delivering native, competitive gaming capabilities. It's an exciting time to cover laptops and AI, as the threads of earlier hardware announcements finally start to come together. It's the anticipated fruit of seeds Nvidia planted with last year's DGX Spark announcement, introducing personal-scale AI devices that are effectively supercomputer hardware. RTX Spark also gives us a glimpse of what we might see from Nvidia's partnership with Intel, as the two chip giants will likely collaborate to bring similar GPU power and unified memory to x86 in the near future. However, looking back only maps the fault lines, and we're watching the ground move now (and again this fall). Here's how I see the laptop landscape shifting around Nvidia's new RTX Spark chips. Laptops Now Have More Competition The biggest, most obvious change with Nvidia's entry into laptop processing is that Intel, AMD, and Qualcomm will now compete directly with Nvidia, turning what was once a two-player competition between Intel and AMD, with a scrappy third fighter taking potshots, into a four-way melee. (That's not even counting Macs, which use Apple's own Arm-based silicon.) That's both good and bad news. On the plus side, more competition means more innovation. The four chip makers will race to outdo each other on features, capabilities, and price. If my economics classes taught me anything, it's that this sort of competition drives down prices and results in better products. The current memory shortage actually highlights this sort of supply and demand, as the market for memory has exponentially increased while the number of suppliers has stayed the same. (I'm flashing back to blackboard drawings of supply and demand curves.) When those pressures come to bear, someone eventually finds a way to meet demand. We're seeing it now with the AI need for local compute, and I'm sure we'll see something similar in memory manufacturing or implementation in the next year or two. Remember the flipside here, however: More players means more fragmentation. Intel and AMD are still cranking out x86 processors, but Nvidia will join Qualcomm in making Arm chips, enabled by Microsoft's own Windows on Arm software efforts. Windows developers will have to either pick a side or work to support both camps. I've found a silver lining, though. Adding Nvidia as a new player actually helps with some of that fragmentation, because it throws Nvidia's sizable weight behind Windows on Arm, which will further the new platform with expansions into gaming and heavy-duty AI. The sheer gravity Nvidia exerts by playing in that space means that more developers will support Windows on Arm, and more refinements will come to the platform. Nvidia's vote for the Arm ecosystem will move investment dollars, developers large and small, and even consumers, who may be more likely to buy a laptop powered by the well-known Nvidia than the less-prominent Qualcomm. The Superchip Era: Is This What We're Calling It? The RTX Spark is an Nvidia product, but it's actually the result of the same Nvidia and MediaTek collaboration we saw with the DGX Spark. Nvidia designed the graphics and AI hardware, but utilized MediaTek's platform integration of Arm Cortex CPU cores, with TSMC handling the physical 3-nanometer manufacturing to physically make the superchip SoC. Back when Intel and Nvidia announced their team-up last fall, I suggested that they would adopt exactly the sort of CPU/GPU+Unified memory approach we're seeing in the RTX Spark. Seeing Nvidia use that exact approach for its own hardware has me quite confident that I'll be proven right on the Intel collaboration doing the same. It makes a lot of sense. That giant SoC approach not only goes way beyond what traditional integrated graphics can do, but it also makes for a larger pool of memory that works for anything, from gaming to AI. As AI features increasingly rely on memory for broader context windows and larger model support, it's a key innovation for allowing more powerful local AI on individual laptops and desktops. We're still in that awkward stage where the new technology doesn't seem to have a proper name yet. Nvidia calls it a superchip, but the industry could coalesce around a new name within the week. With major laptop announcements coming from Computex and new Surface and Windows on Arm details coming from Microsoft Build, everyone in the computing world is talking about this exact technology. The days of awkwardly saying, "It's an SoC, but, like, a more powerful SoC, with unified memory" could quickly give way to a cleaner, more concise term. (I clearly hope so.) RTX Spark is also something of a threat to Apple, with the superchip promising the same sort of power and capability as its MacBook Pro chips. Apple has long had a grasp on the creative class, most recently by making video editors and digital artists choose between Apple's unified memory or Nvidia's CUDA ecosystem. With the RTX Spark, that trade-off shifts dramatically, presenting a non-Apple option with the same sort of developer buy-in and impressive unified memory to enable fast, AI-enhanced workflows. Gaming Grows in Unexpected Ways New Nvidia hardware also points to another happy outcome: More gaming on more devices! With RTX Spark featuring power on par with an Nvidia RTX 5070 Laptop GPU, we now have undeniably gaming-grade hardware in a Windows on Arm device. Here's the thing: We already have some gaming capability with Windows on Arm. If you're willing to settle for playable frame rates in the 30s and 40s, and tinker with things like AMD's FSR, you can game on some of the better Qualcomm Windows machines today. They're also well past low-impact games like Minecraft and World of Warcraft, supporting relatively recent games like Cyberpunk 2077. Regardless, gaming on Arm has faced some major roadblocks. Graphics power is one of them -- the lack of a discrete GPU has left most Qualcomm machines underpowered for modern gaming. Still, the bigger bottlenecks have been the need to use Prism emulation, or worse, many games are simply unplayable due to kernel-level anti-cheat technology that won't run on Arm, until now. Nvidia's stature in the gaming world is hard to overstate. The company has been a mainstay of modern gaming for decades, and the CUDA platform went from being the first GPU compute model to see wide adoption to becoming the backbone for recent advances like DLSS 4. If Nvidia is making an Arm processor for Windows machines, you'd better believe that game makers will start making more Arm-compatible games. In fact, we're already seeing the changes. Microsoft will fix the Anti-Cheat technology on Arm, a detail included in the RTX Spark announcement. The Windows maker also announced that Riot Games and Krafton, developers of Valorant and PUBG, respectively, will bring their game libraries to Arm with native support. Nvidia will also bring DLSS 4.5 to the new RTX Spark laptops. The company specifically named a number of new games in its briefing, showing off hits like Doom: The Dark Ages, Fortnite, and Half-Life 2 RTX, along with demos of Alan Wake 2, Indiana Jones and the Great Circle, and Pragmata. That's all to say that Windows on Arm will soon see a huge gaming upgrade, and I'm sure I'm not the only one saying that it feels a bit overdue. The Biggest Upgrade: Massive AI Power The most significant improvement that Nvidia's RTX Spark will bring to consumer systems is AI power. This is the same server-grade AI muscle that Nvidia has packed into the Nvidia DGX Spark, but optimized for Windows, and with a neural processor (NPU) added to the mix to handle all of Microsoft's Copilot+ features. That makes these new machines the most powerful AI PCs yet, handling all of the on-device features Copilot+ has. These systems will also provide the horsepower to run a 120-billion-parameter open-source model, or run OpenClaw and Hermes AI agents on your laptop. So far, the selection of laptops (or even desktops) that pack that sort of power has been a rarified few -- the DGX Spark, custom-built multi-GPU rigs, and the few laptops with unified memory, like the MacBook Pro and systems with AMD's Strix Halo family. That level of computing is about to get a lot more common, thanks to Nvidia. This shouldn't be a huge surprise. Of all the companies making hay in the AI goldrush, Nvidia has been churning out the digital shovels as fast as physics allows. From new chips to fresh server hardware and new software frameworks, there is no bigger player in the AI sandbox when it comes to providing compute power. With this announcement, it's clear that Nvidia intends to stay at the top, even as locally run AI and agents become the next big thing. Flipping the Power vs. Battery Life Script Qualcomm has made major gains in laptops, in part, by delivering killer battery life. However, the whole value proposition has been that you enjoy better energy efficiency at the expense of true x86 support and peak graphical power. With Nvidia cannonballing into the Windows-on-Arm swimming pool, that proposition changes. Nvidia's name is synonymous with GPUs, and the graphics power on display here is a huge step up from Qualcomm's best. Nvidia's trade-off may still be battery life, however. The company promises "all-day battery" and high efficiency, but that may only hold true for day-to-day productivity. When gaming or heavy, continuous AI use enters the picture, I suspect we'll see much shorter battery times. How short will have to wait for testing to confirm, but the limits of energy efficiency can only be stretched so far when you're powering a GPU's worth of graphics cores or running an AI agent, even when you're not at the keyboard. The Trickle-Down Effect and a Pricing Reality Check We've obviously heard a lot about the heights the new RTX Spark will be able to reach with the full complement of CUDA cores, Grace cores, and 128GB of unified memory. However, those top-specced models won't be the only systems with RTX hardware inside. Manufacturers will have more than one RTX Spark chip to choose from. Nvidia hasn't specified, but pre-announcement rumors clearly mentioned two different classes of chip (code-named N1 and N1X). Manufacturers will be able to configure systems with as little as 16GB of memory, more in line with current consumer systems. That all adds up to lower-powered lower-priced RTX Spark hardware. We may not know what those lesser chips are called, but they're obviously going to exist. Regardless, this is good news. With skyrocketing RAM prices and fierce competition from Intel, AMD, and Qualcomm, we might see Nvidia systems that people can actually afford. For context, the DGX Spark, which has similar specs to the RTX Spark, initially sold for $3,999 for the 128GB model with a 4TB SSD. Retailers actually dropped the price a bit in late 2025, dipping as low as $3,969 through some retailers. Still, eventually Nvidia bumped the price up to $4,699, with Nvidia citing memory shortages as the cause. Top-tier versions of RTX Spark, with the full 128GB of memory, will almost certainly sell for $3,000 or even $4,000. Charging that much for a developer desktop is one thing for professional users investing in cutting-edge AI equipment, but for personal devices, that's a stretch. Even when we compare it with gaming laptops with discrete GPUs, that's a premium most people can't even consider. If Nvidia wants this to be a mass-market product, and not just a niche option for AI developers, content creators, and gamers, it must have affordable versions for the average consumer to buy. The World of Nvidia's Shake Up Whether budget shoppers and mainstream consumers can buy the premium Nvidia RTX Spark laptops that are coming this year almost misses the point. By bridging the gap between desktop-class AI and gaming or creation hardware, Nvidia will force the whole industry to move in several directions. Qualcomm sees both a big boost to Windows on Arm and an incentive to start adding graphics hardware, or risk playing only in the shallow end of the pool with affordable and mid-market machines. Apple's MacBook will face real competition in the AI space as Windows machines can now compete with Apple silicon for local AI models and always-on agents -- a trend that's niche now, but exploding in developer and AI circles. The changes coming for gaming, for Windows on Arm, and graphics in general will similarly spur competitive growth in the laptop world. Like I said at the start, Nvidia's move isn't just a tremor, it's an earthquake. This Computex announcement will cause aftershocks that I expect we'll still feel in the coming years.

Nvidia's RTX Spark could caplitalize where Qualcomm's Arm-based efforts have not -- following the expiration of Qualcomm's Windows on Arm deal, Nvidia stands poised to pick up the slack

Nvidia unveiled the RTX Spark superchip on May 31st ahead of its GTC Taipei event, and right before Computex 2026. The device packs a 20-core Arm-based Grace CPU and a Blackwell RTX GPU on a single package and points it at the one corner of computing where the company has never had a foothold: the Windows PC. The chip carries up to 128GB of unified memory, a claimed 1 petaflop of AI compute, and 6,144 CUDA cores, and it ships this fall in laptops and compact desktops from ASUS, Dell, HP, Lenovo, Microsoft Surface, and MSI, with Acer and GIGABYTE to follow. Microsoft is named as a co-developer, not just an OS supplier, having built new Windows security primitives to run on-device AI agents alongside Nvidia's OpenShell runtime. Branded as RTX Spark, it's the chip the industry has spent three years calling N1X. "For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask, and the PC does the work," said CEO Jensen Huang. Running 120-billion-parameter LLMs with up to 1 million tokens of context, RTX Spark can render 90GB 3D scenes and play AAA games at 1440p above 100 frames per second, all on a chip whose CPU was half engineered by smartphone SoC vendor MediaTek. 'A new era of PC' RTX Spark hasn't come out of nowhere; it's the consumer-oriented sibling of the GB10 Grace Blackwell superchip already shipping inside the Linux-based DGX Spark mini-PC, which carries a price tag currently approaching $5,000 due to memory shortage-related pricing pressure. The GB10 pairs a MediaTek-produced Arm CPU complex with a Blackwell GPU on a TSMC 3nm-class node, joined by Nvidia's coherent NVLink-C2C interconnect and fed by a shared 128GB pool of LPDDR5X. RTX Spark takes that architecture and repurposes it for Windows. We first began to hear about the RTX Spark under its N1X codename back in 2023, when it was reported that Nvidia was developing Arm CPUs capable of running Windows. The chip appeared repeatedly through 2024 and 2025 via the rumor mill, with various delays attributed to factors including Microsoft's slow next-gen work on Arm and soft notebook demand, pushing a planned second-half-2025 debut into this year. For eight years, Microsoft's Windows on Arm program ran exclusively on Qualcomm silicon under a partnership that locked out every other chipmaker. Microsoft chose Qualcomm in 2016, and until the deal lapsed, no rival could ship an Arm chip in a Windows PC. Arm CEO Rene Haas confirmed in an interview in January 2024 that Qualcomm's exclusivity with Microsoft would lapse that year, the first on-record acknowledgment from a principal after years of the deal being treated as an open secret. Reuters had reported in 2024 that MediaTek, Nvidia, and AMD were all building Arm Windows chips to enter once the window opened. Microsoft's role in RTX Spark goes deeper than the Copilot+ certification program it handed Qualcomm, however. The two companies built the agent security stack together at the operating-system level: identity, containment, and policy primitives in Windows, paired with OpenShell's ability to route queries to local models based on a user's privacy rules and to mask personal information in queries sent to the cloud. Satya Nadella, chairman and CEO of Microsoft, said the launch will deliver "unmetered intelligence to every home and every desk with Windows," an outwardly materially closer integration than the Snapdragon X program ever received. Windows on Arm Qualcomm spent its eight years of exclusivity demonstrating that Windows on Arm could work, but failing to make it sell. Snapdragon X laptops moved roughly 720,000 units in the third quarter of 2024, their first full quarter on sale, which Canalys data put at about 0.8% of PC shipments that quarter. ABI Research projected Arm wouldn't clear 13% of the PC market in 2025. Qualcomm's own counter-figures were heavily conditioned: CEO Cristiano Amon's "more than 10%" share claim, made on the company's Q1 2025 earnings call, covered only U.S. retail Windows laptops priced above $800 in a single quarter. A big factor behind this lacklustre performance was software issues. Microsoft's Prism runs x86 apps on Arm, but in our own analysis of Snapdragon X2 Elite Extreme, we found that professional tools like AutoCAD were unsupported, and games crashed or rendered incorrectly under emulation. The original Snapdragon X pitch leaned heavily on battery life as a huge differentiator, but then Intel's Lunar Lake matched that efficiency, giving buyers long battery life on x86 chips that run every Windows app natively, with no emulation and none of the slowdowns or crashes that came with it. Ultimately, Arm's share of Windows never reached the 50% within five years that Arm and Qualcomm had floated back in 2024. Two familiar problems Unlike Qualcomm, Nvidia isn't selling battery life. RTX Spark's USP is the GPU, CUDA, and the 128GB unified memory pool, hardware aimed at local AI, agents, creators, and gamers rather than all-day portability. Adobe is rearchitecting Photoshop and Premiere for the platform with a claimed two-times uplift in AI and editing workflows, and over 100 Windows software vendors, plus game developers including KRAFTON, NetEase, Remedy Entertainment, Riot Games, and Microsoft Xbox are listed as backing the platform. Two problems dogged Windows on Arm for nearly a decade that won't simply disappear with a faster chip, though. First: x86 emulation. Any application without a native Arm build still runs via Prism, and that has meant performance penalties and outright failures across the Snapdragon era. Nvidia's full CUDA and RTX stack is native, which helps AI and graphics workloads, but says nothing about the long tail of legacy Windows software and peripheral drivers. The second problem is Microsoft itself: its slow progress on the next-gen Windows on Arm platform was cited as a primary cause of the N1X delays, and developers won't be getting the full picture of the Windows agent features until Microsoft's Build keynote on June 2nd and 3rd, days after the chip was announced. As for pricing, we've got nothing on that yet. The only reference point is the DGX Spark's $3,999 desktop baseline, a figure that's now approaching $5,000 but also heavily inflated by enterprise networking hardware that consumer-grade devices will omit. That said, LPDDR5X memory costs and TSMC 3nm manufacturing both point toward premium pricing rather than the sub-$700 bracket Qualcomm targeted to broaden Arm's reach. With the RTX Spark, Nvidia is opening a door that Qualcomm could only pry at, carrying the one asset it never had, and inheriting compatibility and OS dependencies that no amount of compute can resolve on its own.

Nvidia's Grace Blackwell superchips are officially coming to the PC with RTX Spark notebooks

COMPUTEX 2026: It only took a year and a half but the same silicon at the heart of Nvidia's DGX Spark AI workstations will soon be powering Windows PCs. During his GTC Taiwan keynote on Monday, Nvidia CEO Jensen Huang revealed the N1X, a high-end mobile processor that combines an Arm-based CPU co-designed with MediaTek with a Blackwell based GPU on board. Marketed under the "RTX Spark" banner, Nvidia's new notebooks and mini PCs signal a deeper push into the a PC arena long dominated by Intel and AMD. But while the PCs are new, the chip powering them isn't. Nvidia was rumored to be working on the N1X for several years now. At CES in 2025 the GPU slinger fanned the rumor mill flames when it unveiled the DGX Spark -- then codenamed Project Digits. The $4,000 AI workstation was powered by a miniaturized Grace Blackwell processor packing 20 ARMv9 CPU cores and a Blackwell GPU with 6,144 CUDA cores, capable of up to 500 teraFLOPS of FP4 compute -- or 1 petaFLOP if you happen to have a workload that supports sparsity. That's fed by 128 GB of unified memory. If that sounds familiar, that's because the N1X and GB10 are essentially the same chip. The liberal use of the term "up to" in Nvidia's marketing does, however, suggest that not all SKUs will have all CPU or GPU cores enabled. The silicon may be the same but the operating system isn't. While Nvidia's DGX Spark and GB10 partner systems shipped with DGX OS, a lightly customized version of Ubuntu 24.04, RTX Spark systems will ship with Windows. This opens the door to high performance mobile gaming using integrated Nvidia graphics. The GPU giant claims the N1X-based systems should be able to manage 100 frames per second at 1440P in AAA games, presumably with the help of AI upscaling tech like DLSS. With up to 128 GB of unified memory shared between the CPU and GPU, RTX Spark systems should be able to handle creative workloads previously limited to high-end workstations. Nvidia suggests top end RTX Spark systems should be able to handle 3D renders requiring 90-plus gigabytes of memory, edit 12K video, generate AI videos, and run 120 billion parameter LLMs with the large context windows required for local agents. As part of the announcement, Huang teased an appearance alongside Microsoft CEO Satya Nadella to discuss the future of AI PCs during the software titan's Build conference, which kicks off tomorrow. As for the hardware, Nvidia has clearly set a high bar for quality. In addition to DGX Spark-style Mini PCs, RTX Spark systems will range from 14 to 16-inches in size, feature aluminum chassis, and color accurate OLED displays with Nvidia G-Sync. The first N1X-based PCs and notebooks from the likes of Asus, Dell, HP, Lenovo, Microsoft and MSI are expected to begin rolling out this fall. We've asked Nvidia to confirm RTX Spark systems pricing, but don't expect the top end variants to be cheap. GB10-based systems ranged from $3,000 to $4,000 at launch. Memory prices have only gone up since then with the DGX Spark now retailing for $4,699. ®

Nvidia's RTX Spark will "reinvent the PC," giving Windows on Arm the huge boost it deserves

* Microsoft and Nvidia have unveiled the RTX Spark (N1/N1X) to supercharge Windows on Arm. * The chip features massive specs: 1 petaflop AI, 6,144 Blackwell RTX cores, 20 Arm CPU cores, up to 128GB RAM. * Surface Laptop Ultra debuts RTX Spark with MPTF and workload scheduling for power, cooling, and peak performance. We recently caught wind that Microsoft, Nvidia, and Qualcomm were planning something big. We then saw a leak claiming that the trio was planning to pull back the curtain on the N1X and N1 chips, which would elevate Windows on Arm to a whole new level. Now, Nvidia and Microsoft have officially shown off the new chip to the world at Computex 2026, and it looks like it'll seriously shake up laptops. Nvidia's N1X and N1 chips are now called RTX Spark It's a game-changer Over on the Windows Blog, Microsoft showed off what the RTX Spark can do. We're looking at 1 petaflop of AI performance, 6144 Blackwell RTX cores, 20 Arm-based CPU cores mdae in collaboration with MediaTek, and support for up to 128GB of unified RAM with a 600GB/s memory bandwidth. Microsoft has also created workload profile scheduling (WPS) to optimize RTX Spark, ensuring Windows leverages all that hardware to its maximum capacity on day one. It manages to pack all of that in a tidy 45-80W power profile to keep it from draining the batter. To keep everything cool, Nvidia enabled Microsoft Power and Thermal Framework (MPTF) on the chip. This will not only ensure that RTX Spark delivers the best performance for your charge, but also keep the internals nice and cool, no matter what you throw at it. And if you still need apps that don't have an Arm-native version released yet, RTX Spark has the 32 and 64-bit x86 Prism emulator "present and optimised" to get them running. RTX Spark is built for agentic use-cases run locally, including Nvidia's own Nemotron models. Huang stated during the company's keynote that Adobe will support RTX Spark, and you will be able to use Adobe's MCP server with your local agent for productivity workloads. Jeff Fisher, senior vice president of personal computing at Nvidia, was understandably very pleased with the collaboration: NVIDIA and Microsoft share a vision that agents are the future of personal computing. RTX Spark combines NVIDIA's full technology stack with Microsoft Windows and is purpose-built for creators, gamers and AI developers in the personal AI era. As well, Jensen shared details of upcoming hardware. There's a DGX workstation built for Windows, packing 768GB of memory, an MSI-based mini PC, and the Microsoft Surface Laptop Ultra. As well, we're already seeing this chip appear in devices from Microsoft, Asus, Dell, HP, and Lenovo, meaning we'll soon have this mighty hardware in our hands. Subscribe to our newsletter for RTX Spark insights Get deeper context by subscribing to our newsletter for analysis of RTX Spark, the Surface Ultra, and what this means for PC-AI, Windows on Arm, and developer workflows. Clear breakdowns, hands-on takeaways, and continued coverage of related hardware and Get Updates By subscribing, you agree to receive newsletter and marketing emails, and accept our Terms of Use and Privacy Policy. You can unsubscribe anytime. Microsoft's Surface Laptop Ultra is the most powerful Surface device to date, debuting RTX Spark to bring high-performance processing in a luxurious package. Alienware makes a big entrance at Computex with the first 5K ultrawide monitor with RGB stripe subpixels It even comes with a "esports mode." Posts By Simon Batt

Nvidia announces RTX Spark as 'the most efficient PC chip ever built'

This fall, Nvidia will officially become a consumer PC chipmaker like Intel, AMD, Apple, and Qualcomm, putting a complete computing chip -- not just graphics -- into the very heart of laptops and mini-PCs. After many months of leaks, it's finally announcing the RTX Spark, the first in a family of chips that will meet or beat the most powerful thin-and-light Windows machines ever, it claims. "This is the most efficient PC chip ever built," says Nvidia senior director of product management Mark Aevermann -- without sharing so much as a single statistic or chart to back that up. The RTX Spark is effectively the same GB10 chip that's in the DGX Spark, the tiny "personal AI supercomputer" that Nvidia released last year, only now it's a family of chips instead of just one. The flagship version appears to be spec-to-spec identical with 20 CPU cores, 6,144 GPU cores and 128GB of LPDDR5X memory. But Nvidia says there'll be lesser versions later, targeting lower prices, and with as little as 16GB of RAM. Like Apple and Qualcomm's chips, this Nvidia chip is Arm-based silicon, meaning legacy Windows software made for Intel and AMD's x86 processors needs to run through an emulation layer to work. That can mean lower performance. But Microsoft has now spent years getting Windows and its Prism emulator ready for Qualcomm and now Nvidia chips, and Nvidia claims its own graphics and AI chops will take the idea further than ever before. With the power of the RTX Spark, Nvidia boasts, you can render a 90GB 3D scene, edit 12K resolution video, or play the graphically intensive Indiana Jones and the Great Circle at a smooth 100fps at 1440p resolution -- all in a 14mm thick laptop without a power cord plugged in. And with up to 128GB of unified memory, tied with AMD's previous gen Strix Halo parts, an RTX Spark laptop or desktop can also host 120-billion-parameter AI agents, something that Microsoft is seemingly excited about for Windows. At Microsoft's Build conference this week, it'll be showing off "new Windows security and containment primitives" that, along with Nvidia's OpenShell runtime, "allows personal agents to run safely and under full user control." Nvidia claims this adds up to "a new personal computing paradigm where AI is the UX" and "users no longer need to master complicated app UIs" because you'll just talk to your PC instead of needing to use mouse and keyboard. Nvidia suggested that, for example, a esports streamer could get their PC to automatically turn off their lights, mute their microphone, and change their broadcasting mode when they want to step away and grab dinner. A designer could use Adobe to automatically turn a sketch into a full image, render a 3D model of it, then create a AI video just by asking. A software developer can automatically monitor their GitHub project and autonomously fix QA issues, with the AI agent taking over the laptop's keyboard and mouse cursor to do "repetitive and boring" tasks. Nvidia says that with the RTX Spark's local AI chops, your data stays private and you won't be burning through tokens to do AI things. I'm not convinced Nvidia has pieced together the Star Trek computer just yet, but it does seem like the company has a lot of partners on board. Almost every major laptop vendor is accounted for, with eight specific laptops already confirmed for this fall: One of those is from Microsoft, which is putting the Nvidia RTX Spark in a new laptop that Surface boss Andrew Hill tells us is "the most powerful thing we've ever made." It's called the Surface Laptop Ultra: Those machines are apparently just the start. Aevermann says Nvidia's partners are already working on over 30 laptops and over 10 desktops, with Acer, Asus, Dell, Gigabyte, HP, MSI, and Lenovo all on board for the latter. "RTX Spark is going to be a family of products that are going to attack a lot of different price points," Aevermann promises. "The overall market opportunity that we see is quite large." And between Microsoft and Nvidia's wrangling efforts, lots of Windows developers are also on board with Arm. The company points out that "Blender, DaVinci Resolve, Maxon Cinema4D, Maxon Redshift, Topaz Photo, CapCut, Cubase, Bitwig Studio, Affinity by Canva and more all run natively on Arm today, as do the audio, video, MIDI, and control peripherals they require." Adobe is on board, with special optimizations for Premiere and Photoshop that take advantage of Nvidia's new chip. Even games with anti-cheat that thumbed their nose at Linux and the Steam Deck are now supporting Windows on Arm, too. Microsoft writes that Riot Games is now bringing both League of Legends and Valorant to Windows on Arm. Krafton is bringing PUBG, and Nvidia tells us it's working with more developers who use Easy Anti-Cheat, BattlEye and Denuvo. (Epic's Fortnite already came to Windows on Arm last November after an announcement last March.) Aevermann says "all the top games will run on RTX Spark and provide a great experience." That's a pretty high bar to meet! Here are some more developers Nvidia says it's working with: There are still many open questions, of course. Neither Nvidia nor Microsoft gave us a clear idea of how much these computers might cost, save that the first batch this fall is "targeting the more premium price points in the market." On "all-day battery life," Aevermann would only say that we should "expect it to be much better than anything you've seen before on RTX laptops" and that "you won't need a charger" if you're not pushing heavy workloads. The chip scales down to "low, low single-digit" wattage and goes as high as 80 watts, he says. The latter means they could theoretically drain bigger laptop batteries in around an hour at full bore. On performance, Nvidia didn't have a single statistic or chart to share, and Aevermann wouldn't answer questions about how the RTX Spark family stacks up to chips from Intel, AMD, Apple, and Qualcomm, saying Nvidia will have more to share closer to launch. But he does say that depending on the application, it has roughly the graphical power of an RTX 5070 mobile GPU, and that we should expect the CPU portion to be "competitive with anything else out there in the Windows space." Nvidia also wouldn't say whether these chips, made on the TSMC 3 process in partnership with MediaTek, are being manufactured in the US or abroad. That got a no comment. Nvidia also wouldn't comment on whether it plans to offer Linux driver support for the RTX Spark, as it's currently focused on Windows. It wouldn't comment on putting the Spark in gaming handhelds, like AMD did with its powerful Strix Halo. But it did answer a question that no, the RTX Spark won't be paired with additional discrete GPUs -- which may limit its potential in desktops beyond the miniature ones, the same way Apple's Mac Pro became limited when its Arm-based chips broke compatibility with discrete GPUs. Maybe it doesn't matter that Nvidia isn't sharing proof to back up its claims. Back in 2020, Apple didn't share any proof when it announced Apple Silicon. But when the M1 arrived, it upended our concept of laptop performance overnight.

Microsoft's Nvidia-Spark-Powered Surface Laptop Ultra Signals a Raw Power Revolution

With 15+ years in tech journalism, Brian has deep experience with PCs, AI, and the intersection of the two. He's also the resident Starlink expert. Alongside Nvidia's keynote at Computex, Microsoft announced a new Surface PC -- the Surface Laptop Ultra -- which will feature Nvidia's new RTX Spark superchip. This Microsoft and Nvidia team-up is the first of several major announcements we expect to see during Computex as the chip maker makes a play to power the next generation of premium laptops, with an emphasis on AI and gaming. The Laptop Ultra expands Microsoft's consumer Surface line with more Windows on Arm technology, following years of Surface built on Qualcomm's Snapdragon X Arm processors. Unlike the Qualcomm chips, though, Nvidia's new RTX Spark superchip combines the processing might of Nvidia's DGX Spark and the graphics capability of an Nvidia RTX 5070 laptop GPU to power a new Windows-based agentic AI paradigm. Slated to launch this fall, the Surface Laptop Ultra bridges the gap between thin-and-light ultraportables and beefy workstation laptops, offering the template for a new consumer category that can drive a full petaflop of AI-ready compute power. It's the next phase of evolution for AI PCs, combining Copilot+ features with the raw muscle to power on-device AI models and agents. Inside the Surface Laptop Ultra The Microsoft Surface Laptop Ultra is built around Nvidia's new CPU, the RTX Spark. This Arm-based superchip features 20 Grace compute cores, paired with 6,144 Blackwell RTX cores. This fusion of CPU and GPU muscle is combined with up to 128GB of unified memory, blowing past the limitations of smaller VRAM caches seen on dedicated GPUs. The system is engineered around the Nvidia silicon, leveraging the high-efficiency hardware to deliver unparalleled power in a slim design. Microsoft says that the Surface Laptop Ultra will weigh less than 4.5 pounds and deliver all-day battery life -- though the company has yet to specify the exact workloads used to calculate that longevity. (Gaming and AI workloads will shorten that estimation considerably, I expect.) Other features are similarly premium, including 15-inch mini LED PixelSense Ultra touch screen, with up to 2,000 nits of peak HDR brightness. Inside is a high efficiency cooling system, optimized for the high compute and full array of CUDA cores inside. Port selection includes HDMI, USB-C, USB-A, headphone, and a dedicated SD card slot, alongside Microsoft's largest haptic trackpad to date. The all-metal chassis comes in two finishes: Platinum and Nightfall. Rewriting Windows for the Age of Agentic AI Microsoft is bringing a lot more to the table than just a spruced up laptop design for the new chip. The Surface Laptop Ultra is the flagship for this new architecture, officially joining the Copilot+ PC lineup, utilizing an integrated NPU in the RTX Spark to support Copilot features. But Windows is also making some big changes to the operating system. Agentic AI is getting kernel-level execution for Windows apps, improving out-of-the-box support and opening up a broader ecosystem for frameworks like OpenClaw. Smarter Windows memory management will handle the 128GB of unified RAM, dynamically allocating resources for heavy GPU and AI tasks as needed. The Prism emulation layer -- which Windows on Arm uses to run x86 programs -- gets new refinements and features thanks to Nvidia's involvement. We don't know the full scope of those improvements, but we won't have to wait long to learn more, since Microsoft Build is happening this week, and we expect the conference to include deep dives into the the revamped OS architecture. Finally, a Thin-and-Light Built for Gaming Gaming and anti-cheat and are both getting native support on Arm, marking a huge improvement for that growing segment of the Windows universe. Combined with the new Prism refinements, even non-native games and legacy creator tools will run smoothly on the new hardware, giving digital artists, developers, and gamers an immediate performance boost. Major gaming studios like Riot Games and Krafton are bringing titles to Windows on Arm platforms, and native, cross-platform anti-cheat software should clear one of the biggest hurdles holding back competitive gaming on ARM-based PCs. Maturing the Vision: 'The Most Powerful Thing We've Ever Made' Microsoft signaled a new era of PCs when it launched the Qualcomm-based Surface back in 2024, calling it the beginning of the AI era. And this new Nvidia hardware signals the maturation of that early vision, offering a new level of power, driving use cases that, even recently, sounded like science fiction. Pavan Dauluri, Executive VP of Windows and Surface, underscored this foundational evolution, noting, "At the end of the day, these devices bring together performance, efficiency, powering advanced local AI workflows in a portable form factor." Cramming that kind of power and agentic capability into something that's lighter and slimmer than the average gaming laptop, and putting into a premium package comparable to a MacBook Pro, is a big step forward in this evolution. As Surface Product Leader Andrew Hill plainly stated, "This is the most powerful thing we've ever made." Release Timing and the Price of Ultimate Power The Microsoft Surface Laptop Ultra will launch this fall, one of several new laptops running on the new Nvidia RTX Spark hardware. Currently, pricing information is not yet available, in part because laptop prices are quite dynamic right now due to the volatility of RAM prices. This will surely play a part in the price of a laptop with 128GB of unified memory. Watch for further news this week as Microsoft Build brings us additional details around the Surface Laptop Ultra and the AI enhancements coming to Windows.

Nvidia unveils DGX Sparrk roadmap for laptops and desktop PCs at Computex 2026 -- three generations outlined, Rubin with LPDDR6 memory, followed by Rosa Feynman

Nvidia is fully committed to transforming Windows on Arm into an agentic AI platform Along with its first-generation RTX Spark platform for desktop and laptop PCs, Nvidia CEO Jensen Huang revealed the company's commitment to future generations of those platforms on its future roadmaps. The company is committed to producing at least two additional generations of Spark platforms for its partners. Beyond the Grace Blackwell RTX Spark chips (the top-end RTX Spark Superchip and an as-yet-undetailed smaller chip), Huang promised that every future generation of the company's platforms will include a Spark chip. That means there will be a Vera Rubin pair of Sparks powered by LPDDR6 memory, and a future Rosa Feynman Spark with a presumably even faster (but as-yet unannounced) memory generation. That multi-generational promise is an important point of trust in Nvidia's commitment to transforming Windows PCs for the agentic AI era. Building a full product and partner ecosystem is a much larger challenge than simply building and shipping a chip. It's clear that Nvidia has a small army of OEM partners ready to take those chips to market and a deep partnership with Microsoft and ISVs to unlock the capabilities of its platforms for Windows and the applications that run on it. In order for those OEMs and software partners to trust that it's worth committing precious time and treasure to Nvidia's platform, Nvidia needs to demonstrate in turn that it's invested in shipping future generations of Sparks, as well. Publicly committing to this roadmap at Computex 2026 is an important demonstration that the Grace Blackwell RTX Spark is just the first step on a longer road rather than a momentary curiosity. Indeed, during our pre-brief call last night, Nvidia was asked why its take on Windows on Arm is different compared to other companies' efforts to create alternative ecosystems to x86 mobile processors from AMD and Intel. Nvidia said that "[it's] investing a lot to make sure the Windows on Arm experience is great," and "the reason RTX Spark can succeed is because our full effort is behind this platform and bringing it to market." Given Nvidia's position as the most valuable company in the world and its leading position not only as an AI hardware provider but a leading developer of high-quality open models and the entire underlying stack to run them, those are fighting words, to put it mildly. And while companies like Apple and AMD are building similar SoCs with powerful GPUs and large memory pools, they lack the broad software foundation that Nvidia has built on top of its products for partners to build with in turn. In addition to these relatively low-power platforms, Nvidia will also be producing Windows on Arm-compatible versions of its DGX Station high-end desktop PC. The DGX Station is built around the GB300 Superchip, which encompasses a 72-core Grace CPU with 496 GB of LPDDR5X memory paired with a Blackwell Ultra GPU offering 252GB of HBM3e and up to 15 PFLOPS of FP4 performance without sparsity. Developers can further expand this system with another RTX Pro GPU over PCI Express. Nvidia also committed to future DGX Stations for high-end Windows AI workstation performance. Although those systems will doubtless be far lower volume products than RTX Spark laptops and desktop mini-PCs, it further cements the company's full commitment to creating a reliable and durable ecosystem for partners to build around and consumers to buy into. We're expecting to learn more about Nvidia's RTX Spark systems and ecosystem this week at Computex 2026. Stay tuned for more details. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Nvidia jumps into PCs with new Arm-based chip debuting in laptops from Microsoft, Dell, HP

Nvidia has emerged as the world's most valuable company by dominating the market for AI chips in the data center. Now the company is expanding its prowess to chips that will serve as the main processor for personal computers, entering an arena that's long been ruled by Intel, Advanced Micro Devices, Qualcomm and Apple. During a keynote address at Taiwan's Computex conference on Monday, Nvidia CEO Jensen Huang unveiled a new N1X processor made alongside Microsoft. It will be incorporated into a new RTX Spark superchip, debuting in the fall on a fresh line of Windows PCs from Microsoft, Dell, HP, ASUS, Lenovo and MSI. "Microsoft and Nvidia are going to reinvent the PC," Huang said. "This is the first completely re-engineered, reinvented line of PCs that has happened in 40 years." Nvidia's initial plan includes more than 30 laptops and 10 desktops with the new chip, a Nvidia spokesperson said. The debut PC processor is made up of two flagship types of Nvidia chips fused together, plus 128 gigabytes of unified memory. It pairs one of Nvidia's Blackwell graphics processing units with the new Arm-based custom N1X central processing unit, custom designed by Taiwanese firm MediaTek. The RTX Spark represents a potentially major shakeup for the PC industry, which is already experiencing significant shifts driven by the AI boom. Arm-based processors like Nvidia's are gaining ground over the traditional x86 processors championed by Intel and AMD, while the overall market for CPUs is exploding into what Huang says will be a $200 billion industry. Nvidia told CNBC in February that CPUs were "becoming the bottleneck" amid surging agentic AI workflows. The next month, Nvidia unveiled an entire rack filled with its Vera CPUs for data centers. While training large models requires mass amounts of parallel math -- excellent work for a GPU -- accessing that data and pushing it out to multiple agents requires more general compute offered by a CPU. Nvidia's new PC processor will be made using Taiwan Semiconductor Manufacturing Company's 3-nanometer technology, currently only available in Taiwan. Anticipation around Nvidia's Arm-based PC chip has been building for years. Reuters reported that the company was working on the PC chip in 2023, as part of a push by Microsoft to get companies to make Arm-based processors for its computers. A Nvidia spokesperson said it's been working on the chip with Microsoft for "many, many years," adding that it will be "far, far more capable, higher performance, more efficiency" than traditional x86 processors. Intel is the original pioneer of the x86 instruction set, debuting it in the 1970s. Intel unveiled its new Xeon 6+ data center CPUs at Computex in Taiwan on Monday. Of late, a flurry of companies have been switching to Arm's alternative power-efficient architecture, which first went mainstream on the original iPhone in 2007. Now Apple makes Arm-based processors for its own computers, launching a pricier line of MacBooks with its latest M5 chips in March. Arm also unveiled its first in-house CPU that same month, and AMD is also reportedly working towards an Arm-based PC chip. The first laptops powered by Nvidia's new chip will be as thin as 14 millimeters, carrying a premium price tag, and will also debut in some small desktop models. While RTX Spark will eventually expand to different price points, Nvidia said it's currently targeted toward creators, AI developers and gamers, "looking for very thin and light laptops, slim laptops, portable laptops, or compact desktops." Nvidia said it will release more performance metrics closer to when the chip hits the market in the fall. For now, RTX Spark is "roughly equivalent" to Nvidia's leading RTX 5070 laptop GPU, according to its spokesperson. Huang also announced at Computex Monday that Nvidia's Vera CPU for data centers is now in full production. Huang said Nvidia is making millions of the CPUs for "a market that never existed before." Vera will be available starting in the fall. Early customers include Anthropic, OpenAI, SpaceX's xAI, Dell, Oracle and CoreWeave. "This is going to be our new major growth driver," Huang said. "These CPUs are going to be both performant, but they also have to be extremely energy efficient, so that we can cram as much CPU as we can into the factory without taking away power from the token generation." "Fast CPUs have become essential to keeping the AI factory moving," said Ian Buck, Nvidia's VP of hyperscale and high performance computing. Buck said that Vera can produce tokens 1.8 times faster than x86 today, "advancing overall agent token performance, enabling smarter, longer-thinking agents and in the end, generating more data center token revenue."

Microsoft unveils Surface Laptop Ultra with Nvidia RTX Spark SoC and up to 128GB of RAM

Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. What just happened? At Computex 2026 in Taiwan, Microsoft unveiled the Surface Laptop Ultra, a 15-inch notebook powered by Nvidia's new RTX Spark SoC. Microsoft is positioning it as a MacBook Pro alternative for creators, developers, AI researchers, and professionals, claiming it is "the most powerful device" it has ever made. The Surface Laptop Ultra features a standard design, without the more experimental elements seen in earlier Surface devices. That means it does not include a removable display like the Surface Book series, nor an adjustable screen that can be moved forward or backward, as seen in the Surface Laptop Studio. The notebook features a 15-inch mini-LED PixelSense Ultra touchscreen display with a 2,880 × 1,920 resolution (262 ppi pixel density) and up to 2,000 nits of peak HDR brightness. Microsoft claims this is the brightest display ever integrated into a Surface device and says it delivers high-precision color accuracy for creative workloads. The laptop is powered by Nvidia's RTX Spark SoC, based on the Grace Blackwell architecture. Built on TSMC's 3nm process, it features an Arm-based 20-core Grace CPU developed in collaboration with MediaTek, alongside a Blackwell-based integrated GPU with 6,144 CUDA cores and fifth-generation Tensor cores for on-device AI processing. The chip delivers one petaflop of AI compute, enabling the Surface Laptop Ultra to run 120-billion-parameter AI models locally without sending data to the cloud. This local AI processing capability is one of the standout features of the new laptop, as it addresses a key concern among privacy advocates and power users regarding how AI features operate on smartphones and PCs. Nvidia says the RTX Spark is designed to be secure from the ground up, utilizing the OpenShell runtime enhanced with Microsoft-designed security and containment primitives, along with support for two widely used open-source and self-hosted AI agent frameworks: Hermes and OpenClaw. The Surface Laptop Ultra can be configured with up to 128GB of unified memory, which Microsoft says can dynamically allocate resources between the CPU and GPU depending on workload. Connectivity options include a full HDMI port, USB-C, USB-A, an SD card reader, and a 3.5mm headphone jack. The device also features the largest haptic touchpad ever included on a Surface laptop. The Surface Laptop Ultra will be available for purchase later this year in Platinum and Nightfall color options. Microsoft has not yet announced the exact launch date, pricing details, or regional availability, but online speculation suggests it will command a premium price, potentially starting at around $2,000 for the base model.

NVIDIA's new chip takes the fight to Apple and Qualcomm

Major PC makers including ASUS, Dell, HP, Lenovo, MSI, Acer, and GIGABYTE are already onboard, with the first RTX Spark-powered systems launching this fall. NVIDIA and Microsoft don't believe the next big upgrade for the PC is a faster keyboard shortcut or a redesigned app. They think AI agents that can do things for you will be the future of computing. At GTC Taipei, NVIDIA announced a new AI superchip for Windows PCs called RTX Spark. It's one of the biggest pushes yet from the company into personal computing, and it comes with a bold claim: the traditional app-centric PC is beginning to give way to a new model where AI becomes the primary interface. That's a radical change from how PCs have worked for decades. NVIDIA envisions users just telling an AI agent what they want done rather than opening apps, navigating menus, and doing things manually. Then the agent would take care of the work on several applications by itself. As NVIDIA described it during the event, "AI is the UX." The implication is that someday, conversations could replace much of the keyboard-and-mouse workflow people rely on today. RTX Spark is the hardware NVIDIA believes can make that happen. The chip integrates the company's AI, graphics, and gaming technologies into a single platform for slim laptops and small desktop computers. It can provide up to 1 petaflop of AI performance, support up to 128GB of unified memory, and run advanced AI models locally. A big part of the pitch is the local processing angle. Running AI agents directly on a PC could offer faster response times, better privacy, and less dependence on internet access. NVIDIA and Microsoft are also developing new Windows features and security tools specifically for these personal AI agents, including NVIDIA OpenShell, intended to allow agents to run safely on user devices. The hardware itself is a big move for NVIDIA as well. The company has long powered gaming PCs with its GeForce GPUs, but with RTX Spark, it's making its most aggressive move yet into the full PC processor market. That puts NVIDIA in more direct competition with established players like Intel, AMD, Qualcomm, and even Apple's increasingly powerful Mac chips. Systems powered by RTX Spark will serve creators, developers, gamers, and AI pros, said NVIDIA. The company says these machines will be able to handle demanding workloads, from local AI inference and software development to 12K video editing, 3D rendering, and high-end gaming. Already, a number of PC manufacturers, including big Windows names like ASUS, Dell, HP, Lenovo, MSI, Acer, and GIGABYTE, are getting devices ready based on the new platform. The first RTX Spark PCs arrive this fall.

Nvidia Now Has a Laptop Chip, and You Can Probably Guess What It's Built for

After storming onto the stage in Taipei, Taiwan, Nvidia CEO Jensen Huang has finally showcased the company's long-awaited laptop-grade CPUs. They're just as promising, and just as perplexing, as we imagined based on nearly a year's worth of rumors and leaks. Huang claimed Nvidia was effectively reinventing PCs with the newfangled RTX Spark platform. He went as far as to state that RTX Spark will handle "every application that Windows has ever run." Huang walked onto the stage holding a PC in both hands, one running 007 First Light and the other Forza Horizon 6. He claimed both titles were running "well." Just how well, we'll have to find out for ourselves. First in this stack is the N1X. The N1X is exciting because of what it represents for the company's GPU architecture. The chip features a Blackwell-series GPU, the same architecture used in the Nvidia GeForce RTX 50-series discrete graphics cards you'll find in many professional or gaming laptops. It uses a 20-core GPU developed in part by MediaTek. The chip supports up to 128GB of unified memory, but beyond these figures, Huang didn't detail what people can expect. But, of course, the point of these chips is to run AI, or at least to run some AI on-device and then rely on the cloud for everything else. Nvidia promised we'll see laptops from practically all the major PC makers, including the likes of Microsoft, Lenovo, Acer, Dell, Gigabyte, and MSI. The N1X will be an ARM-based chip, meaning they use a similar RISC-based microarchitecture to Qualcomm's recent Snapdragon X series. Qualcomm has struggled since it launched its first PC SoC (system-on-chip) to achieve greater compatibility with older x86 apps and drivers. Nvidia may be reaping some of the rewards of Qualcomm's hard work, but it's not all sunshine and rainbows for Team Green. There are still lingering compatibility issues inherent to ARM-based chips, especially when you're working with legacy drivers. Asus has confirmed that its ProArt P14 and ProArt P16 will be among its first models to include the N1X. These devices are made for professional video and graphics work, with options up to 128GB of RAM (on the larger model) and 14- or 16-inch 120Hz OLED displays with 3K and 4K resolution, respectively. One of these Blackwell GPUs on a single SoC may be attractive to the creative sphere. Whether the same chip will be equally impressive for gamers on the go remains to be seen. In that same vein, Nvidia is crafting a whole ecosystem of Spark-type PCs, including more mini PC-like devices built to run agentic software such as Nvidia's own NemoClaw. Huang also showed off a desktop built with a custom SoC that he promised would run Windows. That so-called DGX Station supports up to 760GB of memory and should be able to run a 1-trillion-parameter model on-device. It's certainly interesting to see the world's wealthiest company try its hand at a full laptop processor after putting all its eggs into the AI basket. Moreover, this appears to be a stepping stone toward better graphics and greater efficiency on PCs. That is, if all of Nvidia's claims actually pan out in practice.

Nvidia's RTX Spark finally delivers the MacBook Pro rival Windows needed

Nvidia has unveiled its first laptop CPU for the mainstream, the RTX Spark. It's built to power AI agents, content creation, and gaming, and it's already finding its way into Windows laptops from major brands -- including in a MacBook Pro rival from Microsoft. The just-introduced "superchip" combines Nvidia's 20-core, ARM-based Grace CPU with a Blackwell RTX GPU that includes 6,144 CUDA cores (the same as an RTX 5070), fifth-generation Tensor Cores, and a unified memory system that shares between 16GB to 128GB of RAM. Nvidia claims this delivers up to 1 petaflop of computing power that will help with newly enabled agentic AI support in Windows, including new "primitives" to help securely run agents as well as an OpenShell to help define an agent's capabilities. RTX Spark reportedly has enough power to not only run 120 billion parameter AI models locally (with over 1 million tokens context), but ender very large (over 90GB) 3D scenes, edit movie-grade 12K 4:2:2 video, and play high-end games at 1440p while maintaining over 100 frames per second. Adobe says it's optimizing Photoshop and Premiere to potentially double AI and graphics performance, although it didn't provide a reference point for that figure. Nvidia touts "all-day" battery life, in part thanks to power consumption that can vary from single-digit watts up to 80W. Laptops and compact desktops using RTX Spark will be available this fall from Asus, Dell, HP, Lenovo, Microsoft and MSI. Acer and Gigabyte have promised their own systems later. Microsoft Surface Laptop Ultra: The first RTX Spark PC A premium challenger to Apple Microsoft is the first to share details of an RTX Spark laptop, the Surface Laptop Ultra, and it's not-so-subtly positioned as a competitor to top-spec MacBook Pro models and other premium creator PCs. It will support the chip's maximum 128GB of memory and will supposedly have long battery life even during intensive work. Apple has touted real-world longevity as a selling point, so it's no secret who the audience will be. The Ultra will also have specs that reflect its creative focus, including a 15-inch mini-LED touchscreen with 2,000-nit HDR brightness, the "largest" ever haptic trackpad on a Surface, and a varied port selection that includes USB-C, USB-A, HDMI, and an SD card reader. Microsoft hasn't shared a price and release date for the Surface Laptop Ultra beyond a launch sometime "later this year." However, it's likely to be expensive given that even the more modestly equipped Surface Laptop.8 and Surface Pro 12 start at $1,950. The Ultra is aimed at creatives and gamers who want the best "official" Windows laptop, and are willing to pay for the privilege.

Nvidia's RTX Spark Silicon Brings Supercomputer Ambitions to Consumer Laptops

With 15+ years in tech journalism, Brian has deep experience with PCs, AI, and the intersection of the two. He's also the resident Starlink expert. In a briefing ahead of Nvidia's Computex keynote, we got our first look at RTX Spark, Nvidia's ambitious new Arm-based laptop superchip. Nvidia is actively rethinking the PC paradigm, launching an all-in-one system-on-a-chip (SoC) that combines premium CPU processing power, a gaming-grade GPU, and the core AI architecture seen in Nvidia's enterprise-grade DGX Spark. Designed from the ground up to support autonomous, on-device AI agents, it marks Nvidia's biggest move into the consumer laptop space to date. Whether your PC becomes a high-powered agent or not, these new chips will power some of the most capable laptops ever made. Delivering the petaflops of processing power required to run massive local AI models, the RTX Spark is built to crunch through heavy workloads that are simply too demanding for traditional PCs paired with standard discrete GPUs. We'll surely hear more about it during Computex, but here is everything we know so far about Nvidia's disruptive new silicon. Nvidia Sees a New Era of Agentic Computing In advanced developer circles, the rise of agentic AI has been revolutionary, leading to maxed-out workstation PCs running 24/7 to write code, hunt for system bugs, and coordinate cloud workflows. This hasn't translated into the same level of use in consumer systems, but it's only a matter of time. Nvidia sees agentic AI as becoming the new primary interface for PC users, with conversational tasking replacing traditional app interaction via keyboards and screens. Touting massive benefits like absolute data privacy, tighter security, and eliminating the need to learn the unique interfaces of individual software tools, Nvidia is going all-in on localized AI. Powering that shift? Nvidia's RTX Spark. Inside the Silicon: Meet the RTX Spark Superchip If Spark sounds familiar, that's because Nvidia launched the DGX Spark last year, a Linux machine aimed at developers, powered by Nvidia's Grace Blackwell architecture. That same hardware is now coming to consumers in Windows with the RTX Spark, aimed at a more mainstream audience of consumers, gamers, and developers, but featuring a lot of the same hardware we saw in the DGX desktop. Made with the latest 3nm process, the RTX Spark will combine a 20-core Grace CPU with Blackwell architecture featuring 6,144 CUDA cores. Though not strictly a GPU like dedicated GPUs in current gaming laptops, Nvidia notes that the CUDA core count is on par with the RTX 5070 laptop GPU and expects similar gaming performance. The chip fully supports the entire RTX technology stack, promising smooth 100-fps gaming at 1440p, 3D scene rendering, and up to 12K video editing. In a massive departure from standard PC architecture, the RTX Spark features up to 128GB of LPDDR5X unified coherent memory (OEM configurations will scale from 16GB to 128GB). This massive pool of shared memory allows on-device AI models to run at scales that would completely choke a traditional PC. Nvidia isn't discussing detailed chip specs or direct performance numbers yet, but claims that the chip is engineered to meet and beat anything currently on the market. Not only will these chips be powerful, but they'll also be efficient, scaling dynamically from single-digit wattage for idle web browsing and productivity work, up to 80 watts under full gaming or local AI compilation workloads. Windows on Arm Muscles Up The other detail set to shake up the laptop industry is that Nvidia's chips will run Windows on Arm rather than x86. Much like current Qualcomm chips that run Windows with a combination of native apps and emulated apps (via Prism emulation), the new RTX Spark systems will also run on Arm. Nvidia says it has been working closely with Microsoft to optimize WoA for the new RTX Spark chips, but some of those benefits will carry over to Qualcomm systems as well. Nvidia has been further pushing software vendors to offer native Arm support, optimizing apps for the non-x86 side of things, but even pushing for better PRISM emulation with dedicated Advanced Vector Extensions 2 (AVX2) and anti-cheat support for gaming. Though Nvidia didn't share much about software, the briefing did include information about Adobe rearchitecting several key software tools for 100% GPU-accelerated processing on RTX Spark. From Coding to Creation: Real-World Use Cases The bigger portion of the software conversation revolved around open source frameworks for agentic AI, with out-of-the-box support for tools like OpenClaw and Nous Research's Hermes Agent. Windows is getting new kernel-level support for agentic tools, something we expect to hear a lot more about this week during Microsoft Build. While on-device AI agents could be discussed at great length, Nvidia focused on three example use cases: creative work, developer tasks, and gaming and entertainment. The demo included discussion of using AI for image generation and animation, and of employing Adobe and AI workflows to translate simple mood boards and text prompts into high-quality images and video. Complex animations will leverage tools like DLSS and OptiX. Developers could set agents to the task of monitoring website bugs and proactively finding fixes, which the developer can then choose to implement. Monitoring can extend to code repositories or live web projects, with proposed fixes handled in a sandboxed QA environment to prevent accidents and agent overreach. Finally, Nvidia discussed applying advanced tools like DLSS frame generation to video and 3D rendering. Gaming alongside in-game agents could help optimize hardware settings, manage livestream tasks, or even generate assets for game mods. New DLSS Ray Reconstruction is coming this August, and RTX Spark will support that too. Manufacturers, Availability, and Pricing For all we don't yet know about RTX Spark, we do know when it's coming. Nearly every manufacturer in the PC space will be rolling out RTX Spark laptops and even a handful of RTX Spark mini PCs this fall. Asus, Dell, HP, Lenovo, Microsoft, and MSI were all name-checked. More than 30 premium laptops and 10 desktops will be announced with various configurations and features. We expect several to be unveiled at Computex this week. While Nvidia didn't specify the exact number of chip SKUs, there will be a tiered family of products ranging from mainstream to premium configurations, differentiated primarily by Grace core counts, the number of Blackwell CUDA cores, and memory capacity. Pricing is not currently known. RTX Spark chips are for consumer models (no professional or business versions were announced), but the more premium models will likely align with top gaming and workstation hardware. Prices should be announced in the lead-up to the fall launch.

Nvidia unveils RTX Spark Superchip for laptops and desktop PCs at Computex 2026 - new platform promises to turn Windows into an agentic AI OS with Arm CPU, Blackwell GPU, and 128GB unified memory

Over 30 laptops and 10 desktops coming this fall with "the most efficent platform ever built" Nvidia is transforming Windows into an agentic AI platform at 2026. During his keynote, CEO Jensen Huang revealed RTX Spark: a Windows on Arm platform for laptops powered by the company's RTX Spark Superchip. The company boldly claims that this platform is "the most efficient ever built," and it's throwing its full weight into building a first-class Windows on Arm experience for what it envisions as the next frontier of personal computing. Nvidia says AI agents are already shaping a new mode of interaction with PCs. Instead of relying on the same mouse and keyboard inputs that have defined personal computing for 40 years, the company sees AI agents as a new interface that allows users to command their systems and find information with natural language. And once those agents have their marching orders, they'll need to set goals, call tools, evaluate the quality of their work, and refine it, potentially using local and cloud AI models to achieve those ends. Agents might also continue working on long-running tasks even when the user is away from their system or overnight. That all requires powerful, efficient hardware and lots of local memory. To power all this AI reasoning in the new era of computing it envisions, Nvidia is unleashing the RTX Spark Superchip, a Windows on Arm platform more powerful and capable any other on the market. At full strength, this chip offers up to 20 Arm CPU cores, a Blackwell GPU with 6144 CUDA cores, 128GB of LPDDR5X RAM, and up to 300 GB/s of memory bandwidth. That powerful CPU and GPU, connected over NVLink C2C, and the large memory pool give AI agents and 120-billion-parameter models plenty of power and space for long-running tasks with context lengths stretching to a million tokens, according to Nvidia. RTX Spark will power high-end laptops from partners including Dell, HP, Lenovo, Microsoft, Asus, and MSI. Nvidia says it's worked with those partners to create "the most extraordinary laptops [they've] ever built," with tandem OLED G-Sync displays, "all-day" battery life, premium aluminum chassis with large glass touchpads. Nvidia says that the incredible efficiency of the RTX Spark platform "transforms what a high-performance laptop looks like," so buyers in the promised agentic AI age will no longer need to choose between high performance or thin chassis with long battery life. RTX Spark PCs will also deliver similar performance whether plugged in or unplugged, as we've come to expect from other Windows on Arm and Apple Silicon-powered systems. RTX Spark will also bring this agentic Windows on Arm experience to compact, powerful desktops in the vein of the DGX Spark. In total, Nvidia expects over 30 laptops and "10 or so" desktops to lead the charge when the platform launches. In addition to its agentic AI chops, Nvidia positions the RTX Spark Superchip as a creative and gaming powerhouse. The company promises the platform is good for "100 FPS 1440p gaming," potentially enabled by DLSS 4.5 upscaling and Multi Frame Generation. And its large memory pool means creators can work with massive 3D projects and ultra-high resolution video files like 12K 4:2:2 content without running out of resources. To further the RTX Spark platform's creative chops, Nvidia says it's working with Adobe to rebuild the core of Photoshop, transforming it into a 100% GPU-accelerated application for RTX Spark. Those updates will enable new generative workflows, high-dynamic-range editing, and more natural brushing for artists. And Premiere is also getting a core overhaul that's claimed to enable faster and more sophisticated AI workflows, editing, color, and effects. Adobe will also expose Model Context Protocol controls for AI agents to harness its products. In partnership with Microsoft, Nvidia is also helping to transform Windows into an agentic platform with its OpenShell framework and a "new set of security primitives" that form a set of guardrails, ensuring that local agents and models only have access to the tools and data the user grants them access to. Nvidia says Microsoft will reveal more details of this agentic AI transformation at its upcoming Build conference. RTX Spark systems will begin arriving in the fall of 2026, and we can't wait to dig into them to see whether Nvidia's backing will truly transform the Windows on Arm experience for the agentic AI era - or just make for a really great PC platform. Stay tuned. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Microsoft's Surface Laptop Ultra could be your killer PC upgrade

Expected in 2026, this represents Microsoft's most significant Surface redesign, targeting gaming, content creation, and competing with Apple's AI solutions. Microsoft's surprise launch of the new Surface Laptop Ultra, based on Nvidia's RTX Spark platform, is definitely worth talking about. Everything about this machine suggests an exciting reboot of the Surface lineage. The new 15-inch Surface Laptop Ultra laptop is designed around Nvidia's just-announced N1X and N1 processors, which are the foundation of the Nvidia RTX Spark platform that's been expected for literally years now. Though the RTX Spark platform is similar to Qualcomm's Snapdragon X2 Elite Extreme in that they're both Windows on Arm processors, the highest-end configuration of the N1X family includes a whopping 20 CPU cores and 6,144 Blackwell GPU cores, or up to a petaflop of AI performance. This is GPU power that markedly advances the Windows on Arm game, pitting Nvidia against Qualcomm in a battle for laptop users attention. The Surface Laptop Ultra is the fastest Surface Microsoft has ever made, according to Brett Ostrum, Microsoft's corporate vice president of Surface. The machine should slot in as a replacement to the Surface Laptop Studio 2 and the defunct Surface Book as Microsoft's premium, GPU-rich laptop. The Ultra is designed for gaming, content creation, and yes, massive local AI models (120 billion parameters!) with its 128GB of shared memory. For context, those 6,144 Blackwell cores equal the Blackwell cores in an RTX 5070 desktop GPU. In a laptop, its sits somewhere between the 5,888 cores of a RTX 5070 Ti and the 7,680 cores of an RTX 5080. To date, the most powerful laptop Microsoft has shipped has been 2023's Surface Laptop Studio 2, with an RTX 4070 inside. It's now three years later, and Microsoft has slid back into the laptop race yet, adding graphics and AI to its ongoing conversation about productivity. The Surface Laptop Ultra will ship this fall, according to the company. What makes the Surface Laptop Ultra an 'Ultra?' Microsoft recently launched a refreshed lineup of Surface Laptops for Business. It's a line of 13-, 13.8-, and 15-inch laptops that come with Intel's Core Ultra Series 300 or Panther Lake processors. Aesthetically, the Surface Laptop 8 for Business and its colleagues are deadly dull, representing essentially the same laptop Microsoft has shipped for generations, spanning years. The Surface Ultra represents an entirely new design. "We designed Surface Laptop Ultra from the inside out," Ostrum wrote in a blog post. "Mechanical, electrical, thermal, acoustic, materials, industrial design, and software engineers at the table from day one. The internal architecture and the external form built as one system. Our engineers designed it with the same discipline we know you bring to your craft, where every micron matters and every choice is deliberate." OK, the point is clear: The Surface Laptop Ultra is brand new. The Ultra also solves a fundamental problem with Microsoft's other Surface offerings. If I could highlight some of my nerdier testing, Intel's Panther Lake processor is noteworthy for two things. First, its power efficiency is at the top of the current generation of mobile processors. Second, Intel's claim that its GPU can keep up with a mobile Nvidia GeForce RTX 4050. Both are true. Still, Intel's test platforms backed the Panther Lake processor with a massive 99Wh battery and thick, well-cooled designs. In graphics and gaming, Intel's claims also leaned on its AI frame generation techniques that helped push its benchmark scores higher. In my ongoing testing, Microsoft's Surface Laptop 8 for Business simply can't keep up with other Panther Lake laptops. It doesn't get cool enough to let the Panther chip really roar. My bet, given the weight of the Surface Laptop Ultra, is that Microsoft is now free of those limitations in this new machine. We don't know too much about the Ultra's design, but we know that it's going to be, as the kids say, a "chonky" laptop. It's "less than 4.5 pounds," so it may weigh more than the 3.67 pound 15-inch Surface Laptop 8 for Business. It can also last "all-day" on a single charge, but battery size has yet to be defined. In a brief demo video, Microsoft made sure to show us the dual fans underneath the hood. They also tell us that it's designed for "sustained high performance." Microsoft may be feeling a bit sensitive here. My tests of the Surface Laptop for Business 8 show that CPU performance drops a bit under sustained work. In terms of GPU performance -- the foundation for multimedia apps like Photoshop, Blender, and AI -- GPU performance plunges catastrophically under prolonged load, diving to almost half the available performance. That's directly correlated to the thermal design, in my book. It now looks like Microsoft won't make the same mistake again. There's more. Microsoft has moved to an entirely new display, the 15-inch PixelSense Ultra touchscreen, moving to a mini-LED design for the first time. Some creators look down on a mini-LED because of the possibility of LED "blooming" as individual backlights are lit. But it all depends on how many individual mini-LED backlights there are, and we don't know the answer to that question yet. What we do know is that the display supposedly puts out 2,000 nits of peak brightness using HDR, which Microsoft has never done before. We'll have to see how creators react. (Interestingly, Nvidia indicated that RTX Spark laptops would ship with "color-accurate tandem OLED displays with Nvidia G-SYNC technology, which the Surface Laptop Ultra doesn't have.) Microsoft also showed off the Surface Laptop Ultra with two Thunderbolt ports, a headphone jack, and an HDMI port on one side, and another USB-C port, a USB-A port, and an SD card slot on the other. 128GB of RAM? In this economy? Three years ago, GPU performance would be considered a performance story. Today, with all of Microsoft's work with agentic AI, a powerful GPU has become an everyday assistant, driving agents as they access and modify your files. (Well, Microsoft hopes so, anyway.) On a desktop PC, this isn't that much of a challenge, as those GPUs include their own dedicated VRAM for AI applications. Historically, if you've wanted to run an AI application on a laptop, the available system RAM was split between the GPU and the PC's operating system. Component makers have begun stepping in to save the day. Intel, for example, now lets users decide how much memory gets allocated to the CPU and GPU. That opens up space for larger, more complex AI models. Here, the Surface Laptop Ultra's AI productivity power is decided by two things. First, the size of the pooled memory is enormous. It offers users a shared pool of up to 128GB of RAM, a ludicrous amount amid ongoing shortages in PC memory and storage. Second, Microsoft is working within Windows to deliver "a new higher, smarter limit on total system memory accessible by the GPU." To me, that sounds like Microsoft is taking the work done by Intel and others and formalizing it within Windows, allowing you to tweak your AI and productivity applications how you'd like. In total, Microsoft says that you'll be able to run AI models of up to 120 billion parameters on the Surface Laptop Ultra. Microsoft is also working to increase the size of the memory pages AI accesses, so that continually loading the same token history can be accomplished more quickly, in bigger chunks. Nvidia will be bringing its OpenShell application to Windows, bringing in new Windows security primitives. Hermes Agent and OpenClaw will integrate OpenShell and these new primitives, Microsoft said. Keep in mind that Microsoft's management is seeing Apple's Mac minis (running on Apple's own Arm chip architecture) being snapped up as dedicated boxes to run agentic AI. Microsoft wants a piece of this pie, too. And it's getting a bit greedy: Pavan Davuluri, the executive vice president in charge of Windows + Devices, told me that the size of the shared memory pool means that Microsoft is targeting both agentic AI as well "all the other things that an end user might be multitasking with." No dedicated AI box here -- the Surface Laptop Ultra is designed as a "do it all" PC. Was Qualcomm's Snapdragon just a trial run? Sorry, Qualcomm. It's possible that this has been the moment Microsoft has been working toward with all of its efforts with Windows on Arm. At press time, a report from VideoCardz indicated that the Nvidia N1X contained 20 CPU cores, split equally between Arm's Core X925 cores (which they began showing off in 2024 for AI applications) and the Cortex A-725. There's also a reported 18-core (9+9) variant, plus N1 configurations with 8+4 and 7+3 configurations. Just dealing with the permutations of the Nvidia Arm architecture appears to be why the N1X, and the Surface Laptop Ultra, are shipping in 2026 rather than 2025. Microsoft said that it has worked to enable the Windows scheduler to work in conjunction with all of the cores at once, and to improve the Microsoft Power and Thermal Framework, which routes power efficiently back and forth between the CPU and GPU. From an application perspective, Microsoft also worked with application and games makers to ensure that their code was optimized for the N1X and RTX Spark, and that Microsoft's Prism emulator was tuned to step in and accommodate any issues. Frankly, application compatibility doesn't appear to be the issue here. In my own use of Windows on Arm laptops, I've been able to use pretty much everything flawlessly, save for games. What we don't know are the fundamentals. Aside from the impressive GPU numbers, how fast is the N1X and the Surface Laptop Ultra? How much power will it draw? And, more importantly, how much will you pay? Microsoft has always charged a premium for a Surface device, and 128GB of memory could raise prices to stratospheric levels. Still, one thing is clear. We're talking about Surface again.

Nvidia RTX Spark sounds like it might be a great Windows chip even beyond the AI

Microsoft, Nvidia, and MediaTek are working together on "RTX Spark," a new chip for Windows PCs that's all about AI, but if you look beyond that, it sounds like it might still be a good chip in general. As per usual with any big tech launch in 2026, the messaging around the new RTX Spark chipset is all about AI, specifically about AI agents. Nvidia explains that Spark "powers the world's first Windows PCs purpose-built for personal agents, featuring 1 petaflop of AI performance, industry-leading power efficiency, full-stack NVIDIA AI and graphics technology, and up to 128GB of unified memory" with Microsoft building "a native Windows experience for personal agents." Okay. While the focus is very clearly on AI here, it sounds like RTX Spark is a powerhouse of a chip for Windows PCs in general, delivering massive performance and "all-day battery life" based on Arm architecture. The RTX Spark, which Nvidia calls a "superchip," includes an Nvidia RTX Blackwell RTX GPU and a 20-core Nvidia Grace CPU that was designed in collaboration with MediaTek. There are no specific benchmarks available, but Nvidia says that Spark "can render ultralarge 90GB 3D scenes with OptiX and DLSS, edit 12K 4:2:2 video with the NVIDIA Blackwell decoder, run 120-billion-parameter large language models with 1 million tokens context, and play AAA games at 1440p resolution and over 100 frames per second with ray tracing, DLSS and Reflex." In other words, it sounds like an absolute powerhouse of a chip. While Qualcomm's Snapdragon X offerings have been a huge boost to Windows laptops thanks to great battery life and decent performance, this sounds like a massive leap forward. Adobe apps, a relatively sore point of Snapdragon-based Windows laptops, are getting reworked specifically for this new chip. One of the first Windows laptops using RTX Spark will be the Microsoft Surface Laptop Ultra, which Microsoft has teased with a 15-inch mini-LED display, haptic touchpad, and a ton of built-in ports including an SD card slot and HDMI port. It's coming in "Fall 2026." Other Windows laptops promised to get RTX Spark include machines from Asus, Lenovo, Acer, Gigabyte, and MSI, as well as upcoming HP OmniBooks and a Dell XPS 16 Creator Edition.

Nvidia RTX Spark CPU is now official: "superchip" will power Windows laptops and desktops

Serving tech enthusiasts for over 25 years. TechSpot means tech analysis and advice you can trust. Why it matters: Nvidia just announced what it calls the most efficient PC chip ever built. RTX Spark is a Grace Blackwell system on a chip, 70 billion transistors on TSMC 3nm, with a Blackwell RTX GPU, a 20-core Arm CPU built with MediaTek, and up to 128GB of unified memory. It is purpose-built for agents, and it runs full RTX gaming and creation on the same thin-and-light laptop. It is a genuinely impressive piece of silicon. Ryan Shrout is a longtime technology analyst and industry veteran who has spent over two decades covering PC hardware, graphics, and semiconductors. He previously led technical marketing at Intel and was the founding editor of PC Perspective. He is currently President and GM at Signal65. You can follow him on X @ryanshrout. RTX Spark is also, in its top configuration, the GB10 that is the same chip in DGX Spark, which we have already measured at Signal65. So while we already know where this silicon lands against x86 and against Apple in real CPU and GPU throughput, there are going to be a lot of questions about the competitive comparisons in the Windows laptop space. Start with the chip, then look past it This is known silicon, not a surprise, and RTX Spark is arriving a little later than the original plan, which takes some shine off the raw spec reveal. We are not learning what the chip can do today. We mostly already knew. The GPU still holds up, and it is the part I am most excited about. An integrated graphics processor in the RTX 5070 laptop class, in a chassis as thin as 14mm and as light as 3 pounds, at far lower power, is a real achievement. Nvidia pairs 6144 CUDA cores and 1 petaflop of FP4 with the 20-core Grace CPU over a 600 GB/s NVLink C2C link, and claims near-identical performance whether you are plugged in or on battery. If that unplugged parity holds, it removes a performance cliff that has shaped mobile computing for years. RTX Spark is just a chip, until you remember Nvidia has never won on chips alone. RTX Spark also brings unified memory to the Windows PC in a serious way. It carries up to 128GB of unified LPDDR5X at 300 GB/s, which Nvidia calls the largest GPU-addressable memory ever on an RTX device. This is a console and Apple-Silicon-style design brought to Windows, where the CPU and GPU share one pool. It is what lets a thin laptop run a 120 billion parameter local agent with a million-token context, something a discrete-GPU laptop cannot fit in its VRAM. (Yes AMD Strix Halo did this as well.) The real story is the platform The spec sheet is not why this launch matters. The more important move is what Nvidia is doing with Microsoft. The pitch for RTX Spark is not a faster laptop, it is that your PC becomes an agent. Agents change how you use the machine, from input that now includes voice, camera, and on-screen content, to an interface where you state intent and the agent picks the skills, to compute that keeps working when you step away. This is also where Nvidia brings something Microsoft has not been able to manufacture on its own. Microsoft spent two years trying to lead the AI PC story with Copilot, and it underwhelmed. Asked directly how RTX Spark differs from that Copilot era, Nvidia pointedly declined to defend it and left the judgment to us. The combination of Nvidia silicon, the OpenShell agent runtime, and the Windows platform is a stronger agent story than either company has told alone. The open question is whether this stays an Nvidia story. Will the OpenShell model and the AI-as-the-UX idea extend to Qualcomm, AMD, and Intel silicon, or does the most compelling version of agentic Windows remain Nvidia-aligned? That answer shapes how big this actually becomes. Nvidia enters the Windows PC processor war RTX Spark is the first real consumer (in a very long time at least) Windows PC processor from Nvidia, and it walks straight into a fight with Intel Panther Lake, AMD Strix Halo and its successors, Apple M5, and Qualcomm Snapdragon X. The obvious question is why Windows on Arm works this time when it has stumbled before. The Nvidia answer is the combined weight behind it, Nvidia, Microsoft, the application developers, and the OEMs, plus a CUDA and RTX software stack no competitor can match. Compatibility for older apps runs through the Microsoft Prism emulator, now tuned for the RTX Spark microarchitecture, alongside a deep bench of native Arm software. On the creative side, Adobe is rearchitecting Photoshop and Premiere from the ground up for RTX Spark, tapping the unified memory, the Blackwell GPU, and TensorRT for up to 2x faster AI, editing, and color, and extending both so Windows agents can edit inside them. Blender, DaVinci Resolve, Cinema4D, and even MATLAB, which now officially supports Windows on Arm, fill out a genuinely deep day-one list. For gaming, the full RTX and DLSS 4.5 stack is present day one, with native anti-cheat from Easy Anti-Cheat and BattlEye, so this is not a compromised gaming machine. The lineup This launches with scale. Laptops and compact desktops arrive this fall from Asus, Dell, HP, Lenovo, Microsoft Surface, and MSI, with Acer and Gigabyte to follow. Nvidia has said more than 30 laptops and 10 desktops will come over time from every major OEM, starting premium and reaching many price points. Microsoft putting its most powerful Surface Laptop ever on Nvidia silicon is the headline endorsement in that group, alongside the Dell XPS 16 Creator Edition. What Nvidia did not say, and why it matters There is a catch, though. For all the vision, Nvidia withheld nearly every number that would let you judge it. No comparative performance against Intel, AMD, Apple, or Qualcomm. No game benchmarks. No CPU performance specifics. No Prism emulation overhead, which is the single biggest swing factor for how well older apps and games actually run. No pricing beyond premium first. The keynote is a vision and an architecture, with the proof deferred. The remaining unknowns are shipping-product specifics, not the core silicon, and they are the things worth testing independently. Performance per watt. Unplugged parity. Emulation overhead on real x86 apps and games. Large-model throughput and context handling on device. Real-world performance in Premiere, Photoshop, Blender, and DaVinci. How do these new agentic flows actually work and how broad are they going to be? Battery is the last open item. Nvidia claims all-day battery for productivity and, more notably, near-identical performance on battery, but all-day means very different things to different people and workloads, and there are no specific numbers yet. The real bet On the surface, RTX Spark is just a chip. Impressive, but a chip, the kind of spec sheet the industry ships every year. That read misses the lesson from the data center. Nvidia did not win there on silicon alone. It won on CUDA, on twenty years of developer engagement, and on a willingness to create entire market segments that did not exist until Nvidia decided they should. Accelerated computing, the AI training market, the inference economy, Nvidia did not just supply those, it called them into being and then built the software and the developer gravity to own them. RTX Spark deserves that same lens. The question is not whether this chip is fast. It is whether Nvidia can run the same playbook on Windows, the software stack, the developer pull, and the segment creation, and whether it holds the kind of political power in this industry to make a new category stick where others have failed. No pure silicon vendor has had that sort of leverage over an ecosystem since Intel in the 90s. Nvidia might today. It is not a sure thing. Microsoft controls Windows, the incumbents are entrenched, and the client market has humbled confident entrants before. But if any company can will a new computing segment into existence by force of ecosystem, it is this one. That, more than the petaflop or the memory, is what makes RTX Spark worth taking seriously. Whether it works turns on the things Nvidia has not shown, the pricing, the real value to consumers and developers, and whether the platform stays open or stays Nvidia. Microsoft Build, the day after this news, is the next place we find out. So I will leave you with the question I am sitting with. Is RTX Spark the start of Nvidia becoming a serious force in the Windows PC, or a premium niche the rest of the market routes around? And does your PC actually becoming an agent change how you work, or is that a vision still waiting for proof? This article was originally published on X and is reproduced here with permission.

Nvidia Challenges Apple Silicon With New RTX Spark PC Chip

Nvidia is entering the consumer PC chip business for the first time and has thrown down the gauntlet to Apple, describing its new RTX Spark processor as "the most efficient PC chip ever built." Nvidia says its RTX Spark Superchip is purpose-built to run AI agents that can work proactively across apps and run in the background as a personal "teammate." With the chip, Nvidia says users can "render ultra-large 90GB 3D scenes with OptiX and DLSS, edit 12K 4:2:2 video with the NVIDIA Blackwell decoder, run 120-billion-parameter large language models with 1 million tokens context, and play AAA games at 1440p resolution and over 100 frames per second with ray tracing, DLSS and Reflex." The chip was announced by Nvidia chief executive Jensen Huang at the Computex conference in Taipei on Monday. It's a big play for a company traditionally focused on graphics cards to move into the kind of integrated silicon that runs an entire laptop. It also puts the RTX Spark on a collision course with Apple's M5, widely regarded as the laptop chip to beat for running AI tasks on-device. Like Apple's chips, the RTX Spark is Arm-based, pairing an Nvidia Blackwell RTX graphics processor with a Grace CPU. It's effectively the same GB10 chip that's found in the DGX Spark, the tiny "personal AI supercomputer" that Nvidia released last year. Microsoft's new 15-inch Surface Laptop Ultra will be among the first machines to ship with the integrated silicon. The machine features a mini-LED touchscreen, the largest haptic touchpad Microsoft has fitted to a Surface, and a selection of ports covering HDMI, USB-C, USB-A, SD cards, and headphones. Microsoft says the Surface Laptop Ultra will arrive later this year. Pricing has not been announced, but Nvidia has suggested the first wave of RTX Spark machines will target the premium end of the market.

Nvidia RTX Spark is here, and it's going to change laptops forever -- here's everything you need to know

One of the worst-kept secrets of Computex is official, as Nvidia's CEO Jensen Huang just announced RTX Spark Super Chip family -- Team Green's first-ever all-in-one laptop silicon that takes on Intel, AMD, and Qualcomm in a big way with a GPU equivalent to an RTX 5070. Set to release this fall, this feels like just another ordinary ARM chip. It's an architecture Team Green's used to with the likes of the Nintendo Switch 2's chip or the one that powers its DGX Spark mini supercomputer. This guarantees rapid speeds with effective power efficiency. But the Nvidia twist here is that it's packed with CUDA cores, which is sure to make that integrated GPU one helluva gaming monster, and it's set to change the way you interact with a Windows PC forever with an agentic twist. I will be testing it extensively during my time out here at Taipei, and talking to a couple of the engineers behind it. But for now, I've taken a deep dive into the specs tables and found what could be a breakthrough moment similar to when Apple brought the M1 to the world back in 2020. Nvidia RTX Spark Super Chip specs Nvidia RTX Super Chip devices Nvidia is expecting a huge lineup of systems to launch with RTX spark -- with plans to release over 30 laptop models and well over 10 desktop variations, targeting a range of price points. But as of now, there are a whole lot of devices launching this fall with the RTX Spark Super Chip inside. On the laptop side, we have: * Asus ProArt P14 and P16 * Dell SPX 16 * HP OmniBook * Lenovo Yoga Pro 9 * Microsoft Surface Ultra * MSI Prestige N16 These devices will be as thin as 14 millimeters, and as light as three pounds -- a serious upgrade over the chunky Nvidia gaming laptops you're used to. But they're not stopping at laptops. In an announcement that probably woke up Anthony Spadafora (our mini PCs king) from his nap, RTX Spark will be coming to small form-factor desktops that are "small enough to disappear on a desk." These will be made by Dell, HP, Lenovo, Asus, MSI, Acer, and Gigabyte. Explain it with pizza This is partly inspired by the tasty 'za I found while typing this, but also shoutout to Nvidia for specifically pointing out that I just "love to explain stuff with pizza." You asked for it, and now you're getting it! In the laptop space, you've got various ways of serving pizza. * The standard meal (an x86 SoC): This is like ordering a plain cheese pizza. It works perfectly fine for everyday eating, but if you want heavy flavor (think intense gaming or AI generation), it lacks all the necessary toppings. Don't get me wrong, Intel and AMD have been adding some new garnish, but they're still a little hampered by architectural limitations. * The combo meal (x86 CPU + dedicated GPU): For that premium flavor, you have to order the standard pizza plus a massive side of premium toppings. Passing these ingredients back and forth across the table does the job, but it takes up way more space (think thicker gaming laptops) and requires a whole lot of power. * The efficiency slice (Snapdragon X Elite): Qualcomm is already dealing in making efficient single-crust pizza using a similar recipe (Arm architecture). It's great for saving power, but Nvidia's world-famous secret sauces (CUDA and RTX) can't be matched. So in this backdrop, I can only describe the RTX Spark Super chip as a "super pizza." Instead of ordering a meal with lots of separate sides, this is the all-in-one slice baked onto a single, highly efficient crust. Half of the pizza is that custom CPU, and the other half is loaded with incredibly spicy, high-end toppings in that Blackwell GPU. But the magic is in the cheese (the memory). Instead of dividing the cheese between the crust and toppings, Spark uses up to 128GB of "unified cheese" that the entire pizza shares instantly at blazing speeds (five times faster than standard PCIe Gen 5). And because it's baked so efficiently, you can take this super pizza anywhere without it getting cold (all-day battery life). ...am I stretching this analogy? Absolutely. Do I regret it? No (hi Nvidia if you're reading this). Going agentic With this increase in RTX Spark's brute force, Nvidia is working with Microsoft to shift personal computing on Windows from a passive tool to a proactive agent. This is a one petaflop AI superchip... For context, some quick math I did while sitting in the keynote, the AMD Ryzen AI Max+ 395 can do around 60 Teraflops. This thing is built differently. To go back to the pizza analogy, not only can it make a mean slice, but it can probably feed it to you, too. It can work across apps on your device -- stringing together different tools like any user would to get the job done. For example, an artist can provide a sketch and a mood board, then use Photoshop not only to generate the art locally but also to turn it into a 3D image and animate it into a video. And this is all done without ever needing cloud compute. Or if you're a gamer, you can use G-assist to tweak critical settings like monitor optimization, or (if you're streaming) tell it to operate your lights, mute the mic, and turn on your be right back scene in OBS. Essentially, it can take complete control of your system -- even the mouse and keyboard to navigate as a human user would. Keeping this all secure is Nvidia OpenShell, which brings enhanced security to ensure your model use works locally and privately. It can even mask personal information before ever using a cloud-based AI model to do anything. For the players But of course, there are big gaming capabilities here too. This is Nvidia, after all. Team Green says this falls into the same performance class as an RTX 5070 laptop GPU but uses significantly less power to reach that level. That means you could be playing the likes of Cyberpunk, Doom, or Indiana Jones at 100 FPS at 1440p resolution. And given its ARM, the RTX Super Chip is capable of this level of performance both when plugged and unplugged. Of course, battery life will vary based on how much you throw at it, but when I asked Nvidia, a tweak of maximum frame rate and graphics settings can optimize this nicely for long gameplay sessions on flights. Now, what about the emulation part of it? General apps are included in this, too, but gaming has been the big test for Windows on Arm, and that Prism emulation layer used to translate apps written for other CPUs to this new architecture. Well, Nvidia's thought of that too. The company is working with a large list of developers to either port games natively to Windows on Arm or ensure they run well through the emulator. And with that hefty GPU, anything that's particularly taxing on that will have almost no negative effect on performance. Plus, Nvidia has collaborated with all first- and third-party anti-cheat software providers to ensure competitive multiplayer games launch without a hitch. Nvidia RTX Spark vs Apple M silicon vs Snapdragon X2 Elite So, since I'm calling this Nvidia's M1 moment, how does it stack up with Apple laptop silicon? While I'm at it, I'm gonna throw Snapdragon into this comparison too, as they're all similar-ish in Arm architecture. But the crucial differences show that it's not about the tools you have, but rather how you use them. You see, Apple silicon isn't just about the brute force of the hardware; it's also about its tight integration with software that extracts that performance. In gaming, for example, the MetalFX layer has enabled MacBook Pros to do some pretty great things, like run Cyberpunk 2077. And while Snapdragon X2 Elite is level pegging in certain benchmarks, there are still some moments where it can feel like a bit of a blunt instrument against Windows 11 -- especially in that GPU department. But with RTX Spark, this is Nvidia we're talking about. That integrated GPU supports DLSS 4.5, so you can bet your bottom dollar that the out-of-the-box gaming performance is going to be clearly superior to Apple and Snapdragon. On top of that, Team Green and Windows are rewriting the rules on local agentic AI, so you're getting a heap of AI coworkers too, with those CUDA cores for the raw brunt. Nvidia RTX Spark vs Intel Core Ultra Series 3 vs AMD Ryzen AI 400 series Why did I put these separate from the face-offs with Snapdragon or M5? Well, it comes down to the different ways these chips compute stuff. Nvidia, Apple and Qualcomm all have Arm architectures, which do things very differently to the x86 chips from Intel and AMD. That being said, the 30+-year-old x86 architecture does bring some material benefits over ARM -- namely, zero app compatibility issues whatsoever. It's a challenge that Microsoft and Nvidia are overcoming with a separate Arm-specific version of Windows 11, alongside working closely with developers. That could narrow the library you can play! However, for what it may lack in this area, it absolutely makes up for in a sweeter blend of performance and power efficiency. While x86 is all about multitasking its way to doing things you ask it to (not the best on that battery), ARM tackles one task at a time very quickly. That's easier on the power while bringing high speeds to the table, too. Outlook It's been a long road since the first time we heard rumblings about this chip. In fact, I even put this in my Computex 2025 predictions! But now it's here, I think it's fair to say that RTX Spark is far beyond any expectations I had. Not only is it a beastly integrated GPU for monstrous performance and impressive power efficiency (especially in gaming), but the software tie-ins fundamentally change how a Windows laptop works with you. This is going to be an interesting few months as we count down to the fall launch. Now the wait is on for an Nvidia RTX Spark gaming handheld...that would slap so hard! Follow Tom's Guide on Google News and add us as a preferred source to get our up-to-date news, analysis, and reviews in your feeds. Subscribe to Tom's Guide on YouTube and follow us on TikTok. Finally, you can visit our dedicated Tom's Guide Savings Squad hub for expert help on getting the best products for less.

Nvidia's RTX Spark chip powers the Microsoft Surface Ultra AI laptop

Step aside, basically every other laptop out there. Credit: Microsoft Microsoft's got a new Surface coming, and it's the most powerful laptop the company's ever made. Announced at Computex 2026, the Microsoft Surface Ultra comes with an Nvidia RTX Spark, a new all-purpose PC chip from the graphics chip giant. In a press release, Microsoft called the Ultra all sorts of otherworldly superlatives, calling it a laptop for "world makers," but what that translates to is an extremely powerful workhorse laptop capable of running complex (you guessed it) AI tasks. On the surface (sorry), the Surface Ultra is a pretty regular 15-inch laptop designed for professionals, with a mini-LED display, a ton of ports, and an extra-large touchpad. It weighs under 4.5 pounds, and has "all-day battery life," says Microsoft, though not all specs are available at this point. What's inside is what makes this one interesting. First, there's the RTX Spark, Nvidia's new chip for PCs. It appears to be a variant of the GB10 chip, which powers Nvidia's miniature DGX Spark supercomputer announced in January 2025. It comes with 6,144 Blackwell GPU cores, 20 CPU cores, and offers 1 petaflop total of AI performance. You can configure the Ultra with up to 128GB of RAM, making it a portable AI supercomputer (that's also pretty great at regular office and gaming tasks). In AI terms, the Microsoft Surface Ultra should be able to run 120 billion-parameter models locally. For reference, that's on par with a Mac mini equipped with 128GB of RAM. There's no word on pricing, and the Surface Ultra isn't arriving until the fall. It's not the only Nvidia RTX Spark computer that's coming, though. For this one, Nvidia has partnered with basically every Windows PC maker under the sun, including Dell, Asus, HP, and Lenovo; those are all also coming this fall.

Microsoft debuts Nvidia-powered Microsoft Surface Ultra laptop

Why it matters: The move, previously reported by Axios, puts the hottest name in chips behind Windows as Microsoft tries once again to redefine the PC for the AI era. Driving the news: The new Microsoft Surface Laptop Ultra will be powered by the RTX Spark, a PC chip that is similar to the one that powers the Nvidia Spark line of AI desktop machines, per The Verge. * Microsoft said the device, which has a 15-inch mini-LED touchscreen, can be outfitted with up to 128GB of unified memory and can run AI models of up to 120 billion parameters. * Nvidia-powered Windows laptops and desktops are also coming this fall from Asus, Dell, HP and Lenovo, among others. * Nvidia says the laptops will have screens from 14 inches to 16 inches and can be as thin as 14 millimeters. What they're saying: ""The PC is being reinvented," Nvidia CEO Jensen Huang said in a statement. "For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask -- and the PC does the work." * "The work from creators, developers and AI builders has a common shape: massive scenes, long compile cycles, local models and datasets that no longer sit politely in the background," Microsoft corporate VP Brett Ostrom said in a blog post. "We built Surface Laptop Ultra to meet that work without flinching." Yes, but: Microsoft and Nvidia were short on the big details. Microsoft said only that its device would come later this year, with no indication of price, though it's expected to be a very high-end device. * Nvidia said that Windows laptops with its chips would start shipping this fall. The big picture: The launch of the new Surface comes amid a major PC tradeshow in Taiwan and just ahead of Microsoft's Build developer conference later this week in San Francisco. * There Microsoft will once again look to position itself -- and Windows in particular -- as at the center of the AI revolution. * Microsoft's first effort at an AI PC, the Copilot+ PC, was marred by a series of setbacks, including a lengthy delay and security concerns over its signature feature, Recall. * However, the move toward agents that can automatically perform tasks on local PCs has provided what it sees as a fresh opening. * The company has been embracing OpenClaw since earlier this year, creating a new team led by veteran coder Omar Shahine. The company also has OpenClaw founder Peter Steinberger (now employed by OpenAI) scheduled to host a breakout session at Build. * Microsoft is expected to unveil new software to help make Windows a hub for agentic work at Build. The bottom line: While most AI work has been done in the cloud, Microsoft's push to have things run locally could find newly receptive ears.

Nvidia's New Chip Aims to Upend the Creative Laptop Market

Microsoft and Nvidia made joint announcements today. Microsoft is launching a brand-new Surface Laptop Ultra, the most powerful Surface Laptop ever built, and it is powered by Nvidia's new RTX Spark system-on-a-chip, a "new superchip that reinvents Windows PCs for the era of personal AI agents." Nvidia RTX Spark Promises High-End Performance With Laptop-Friendly Efficiency Nvidia describes RTX Spark as designed for "AI, creating, and gaming". It promises it delivers both power and efficiency, making it ideal for ultra-thin, lightweight laptops like the Microsoft Surface Ultra. "The PC is being reinvented," says Jensen Huang, Nvidia's founder and CEO. "For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask -- and the PC does the work. RTX Spark brings everything NVIDIA has built -- CUDA, RTX, our AI platform -- into a single superchip. Local agents. Frontier models. Creative workflows. RTX games. All on a laptop. This is the new PC. The personal AI computer." While there are many AI-related buzzwords here, there is nonetheless interesting hardware on offer. The RTX Spark features an Nvidia Blackwell RTX GPU with 6,144 CUDA cores and fifth-generation Tensor Cores. It features a high-performance, efficient 20-core Nvidia Grace CPU. Nvidia worked with MediaTek on a custom CPU design, promising "best-in-class power efficiency, performance, and connectivity." Microsoft's Most Powerful Surface Laptop Ever Essentially, Nvidia (and Microsoft) want a piece of the lightweight but powerful laptop space that Apple has dominated in the Apple Silicon era. Microsoft notes that it has targeted performance, efficiency, and style across the board with Surface Ultra. It is Microsoft's first laptop to include an Nvidia Blackwell RTX GPU with up to 128GB of unified memory and full CUDA support, for starters. It also promises to be quiet, efficient, and pretty. "Uncompromising craft meets raw power. Built for world makers who demand cutting-edge performance and portability," Microsoft promises. To that end, the notebook features an all-new thermal system designed for heavy, professional-grade workloads. Microsoft says the new thermal system has up to 2.5 times the thermal capacity of the Surface Laptop 7th edition 15-inch. The 15-inch laptop features a mini-LED PixelSense Ultra touchscreen with a 3:2 aspect ratio, 262 PPI resolution, and up to 2,000 nits of peak HDR brightness. Microsoft also touts its color performance, saying the screen is designed for "pros who make critical color and exposure decisions." The Surface Ultra Laptop promises all-day battery life, a compact and pocketable charger, and a 30% larger touchpad with haptic feedback. The laptop boasts a "full set of maker-friendly ports," including USB-C, USB-A, and HDMI, plus a headphone jack and SD card slot. While both Microsoft and Nvidia are emphasizing the RTX Spark's AI capabilities, it is important to consider that this matters beyond the agentic AI features Microsoft is adding to Windows. It also matters beyond AI-powered vibe coding or generative AI apps. Many of the creative applications photographers and videographers use every day, including Photoshop, DaVinci Resolve, and others, rely heavily on AI for many of their latest features. Many creative apps use AI for masking, for example, and those rely on neural processing power. It's easy to dismiss much of the AI hype as irrelevant to those who actually create art in the real world, but AI is increasingly part of creative software. Adobe has already publicly committed to fine-tuning its creative apps, such as Photoshop and Premiere, specifically for RTX Spark. Lots of other companies are on board, including Blackmagic Design, CapCut, Filmora, Blender, Topaz Labs, and more. "The best creative work in the world happens in Adobe tools from Adobe Firefly to Photoshop and Premiere, and the expansion of our partnership with Nvidia and Microsoft will make those experiences faster and more powerful than ever," says Shantanu Narayen, chair and CEO of Adobe. "Together, we are building AI-native creative experiences for RTX Spark that deliver the performance, intelligence and responsiveness people need to create at the pace of their ambition." RTX Spark Is Coming to More Than Microsoft Surface Ultra While the big news today focuses on Nvidia's RTX Spark and the Microsoft Surface Ultra, Nvidia explains that RTX Spark is not exclusive to Microsoft's upcoming Surface laptop. The RTX Spark will also feature in Asus ProArt P14 and P16 laptops, the Dell XPS 16, HP OmniBook models, Lenovo Yoga Pro devices, and MSI Prestige N16 laptops. All of these RTX Spark-powered notebooks are currently slated to arrive this fall, so many details remain unknown, including full configuration specs and pricing. However, Nvidia says its RTX Spark will feature in machines across a diverse range of price points and enable thinner, lighter, and more efficient laptop designs. The era of the chunky, load, LED-laden gaming PC laptop may very well be over. Apple's MacBooks have long won on style, but initial reveals of RTX Spark-based machines suggest the aesthetic gap between Mac and Windows may be closing.

Watch out, Apple - Nvidia just unveiled its RTX Spark Arm 'superchip' to take on the M5 at Computex 2026

Nvidia has just announced its new Arm-based laptop chip at Computex 2026, firing a warning shot across the bow of Apple (who has found great success with its own Arm-based M-series chips), as well as Intel and AMD. While Nvidia has been mainly associated with graphics cards, and AI, in the past, the announcement of its RTX Spark chip, which will power future Windows 11 laptops, could be a real game-changer. While there are a growing number of Windows 11 laptops running on Arm chips, primarily from Qualcomm, the fact that Nvidia, one of the biggest companies in the world, is throwing its hat into the ring is certainly exciting. As Nvidia claimed ahead of the reveal, alongside Microsoft and Arm, a 'new era of computing' has begun. Despite working closely with Qualcomm on Arm-based Windows 11 laptops, Microsoft has failed to match the success that Apple has had with its modern Macs, which ditched Intel processors back in 2020 for its incredibly popular M-series chips, including the latest M5 variant. Nvidia's entry into laptop CPUs could prove to be a major shake-up of the industry - and it couldn't have come at a better time. Game on? The Nvidia RTX Spark comes with 20 CPU cores (the CPU has been custom-designed by Nvidia and MediaTek) and 6,144 CUDA cores based on Blackwell architecture. Talking to representatives of MSI, one of the first hardware manufacturers to make an RTX Spark-powered laptop, this means the integrated GPU is roughly equivalent to an Nvidia RTX 5070 GPU. That's a pretty exciting proposition, and could pave the way for exceptionally thin and light gaming laptops in the future. For the time being, however, it's important to note that RTX Spark laptops won't be aimed at gamers, but rather content creators. That's because this is still an Arm-based chip, and native compatibility with PC games remains scarce without an emulation layer such as Prism, which might allow PC games designed for traditional Intel and AMD hardware to run on Arm, but has an impact on performance. Instead, RTX Spark laptops will be aimed more at content creators, and Dell, MSI and Lenovo are major laptop makers that will be making those laptops. Rumors suggest the Nvidia RTX Spark will have a TDP of 45 - 80W, and because the chip includes both CPU and GPU, I'm a bit concerned that this could mean the N1X struggles in pure gaming performance. A lot will ride on how power-efficient the chip is. The RTX Spark will also support up to 128GB of LPDDR5X memory, and can use Nvidia's gaming tech, such as DLSS upscaling, ray tracing effects and G-Sync. What about the rumored Nvidia N1? Rumors had suggested that Nvidia would also announce a lower-powered chip witheither 12-core (2,560 CUDA cores) and 10-core (2,048 CUDA cores) configurations and supports up to 64GB of LPDDR5X memory, but this wasn't announced at the keynote. Laptop makers team up Jensen Huang also showed off RTX Spark-powered laptops on stage from MSI, Lenovo, and other major laptop makers, with a promise of these landing in 'the fall' - so don't expect anything before September, I guess. These laptops will be thin and light, with some sporting tandem OLED screens with G-Sync. These will be premium laptops, and this is perhaps my biggest concern: how much will RTX Spark laptops cost? No details have been released, but they could end up being very expensive. This will likely limit their appeal and popularity, and I hope it doesn't mean a return to Windows on Arm laptops that cost a fortune, which is what happened a few years ago. If these laptops are more expensive than M5 Max-toting MacBooks (which launched at $2,199 / £2,199 / AU$3,499), then they are going to struggle. Time will tell, then, how scared Apple will be by Nvidia's new RTX Spark chip. Jensen Huang will join Microsoft's CEO, Satya Nadella, on stage at Build 2026 tomorrow, June 2, where he'll go into more detail about RTX Spark.

Nvidia's RTX Spark chip targets the Mac Studio, with Asus and MSI calling the first dibs

Nvidia just built a tiny AI box that wants to fight the Mac Studio This story is part of our coverage of Computex, the world's biggest computing conference. Updated less than 3 minutes ago The battle for compact high-performance desktops is heating up, and Nvidia appears ready to enter territory long dominated by Apple's Mac Studio. At Computex 2026, MSI unveiled a new AI-focused mini PC called the MSI EdgeMesa N AI, powered by Nvidia's brand-new RTX Spark platform. The launch signals Nvidia's growing ambition to push AI computing beyond traditional gaming desktops and into compact creator and workstation machines. More importantly, it also shows PC brands moving aggressively toward Apple's increasingly successful formula of powerful desktop performance inside small, minimalist systems. A tiny AI workstation built around Nvidia's new RTX Spark platform MSI's EdgeMesa N AI is one of the first mini PCs announced using Nvidia's new RTX Spark chip architecture. The system is designed specifically for AI workloads, local generative AI applications, creative software acceleration, and edge computing tasks. Recommended Videos While MSI has not fully disclosed every hardware detail yet, the company confirmed the mini PC combines Nvidia RTX Spark graphics with Intel-based processing hardware inside a compact chassis aimed at creators, developers, and AI-focused users. The system is being positioned less like a traditional gaming PC and more like a local AI workstation capable of handling generative AI models, accelerated creative tasks, and productivity workloads directly on-device. That positioning immediately invites comparisons to Apple's Mac Studio, which has become increasingly popular among creators, video editors, and developers looking for desktop-class performance in smaller form factors. MSI is not alone either. Other PC manufacturers, including ASUS, are also expected to adopt Nvidia's RTX Spark platform for their own compact AI-focused desktops. The broader trend reflects how quickly AI computing is becoming a central selling point for consumer hardware. Instead of relying entirely on cloud services, companies increasingly want local AI processing for privacy, lower latency, and offline functionality. MSI says the EdgeMesa N AI is designed for local AI inference, AI-assisted workflows, content creation, and advanced multitasking scenarios that traditionally required much larger desktop systems. Why this matters For years, Apple largely dominated the premium compact workstation category with devices like the Mac Studio and Mac mini. Now, Nvidia, alongside major PC brands, appears ready to challenge that space directly. The RTX Spark platform represents Nvidia's attempt to create a standardized AI-focused desktop ecosystem for Windows PCs, particularly as AI workloads become more important for creators, developers, researchers, and businesses. The shift also highlights a much larger industry transition happening right now. AI acceleration is rapidly becoming just as important as traditional CPU and GPU performance in next-generation PCs. What happens next MSI has not yet confirmed pricing or final availability details for the EdgeMesa N AI. However, the company is expected to reveal more specifications and launch timelines later this year. As more manufacturers adopt Nvidia's RTX Spark platform, compact AI desktops could quickly become one of the biggest new hardware categories emerging after the generative AI boom. The bigger question is whether Windows-based AI mini PCs can truly compete with Apple's ecosystem advantage and silicon efficiency. But one thing is becoming increasingly clear: the fight for the future of desktop computing is no longer just about raw performance. It is increasingly about who can build the smartest machine in the smallest possible box.

AppleInsider.com

Nvidia has stepped into the processor market with its RTX Spark, but at first glance, it's clearly behind Apple Silicon by a considerable margin. Computex 2026 is underway, and Nvidia has formally stepped into the processor ring with its own chip. Nvidia calls the RTX Spark a "superchip" for Windows PCs that have massive AI performance. This chip consists of an ARM-based Nvidia Grace CPU with 20 cores, as well as an Nvidia Blackwell RTX GPU with 6,144 CUDA cores. There's also fifth-generation Tensor cores, up to 128GB of unified memory, and a 600GB/s Nvidia NVLink-C2C interconnect providing high-bandwidth communications between the elements. According to Nvidia, it is "designed for AI, creating, and gaming," with the intention of it being used to help create slim Windows notebooks with all-day battery life, but massive performance capabilities. This includes rendering massive 90GB 3D scenes for games, generating 4K AI video, and 12K video editing. More handily for AI researchers, it will also be capable of running a 120 billion-parameter large language model with up to a million tokens of context, using local agents. "The PC is being reinvented," said Nvidia CEO Jensen Huang, referring to users predominantly launching apps and manually doing work. Instead, RTX Spark is made to enable "local agents, frontier models, creative workflows, RTX games" on a notebook. "This is the new PC," he declared in a press release. "The personal AI computer." Apple Silicon-esque Undoubtedly, this is a big move for Nvidia, and a major chip introduction that can dramatically affect the Windows notebook market in general. However, Nvidia is still working to catch up to Apple. Nvidia's chip shares many of the same core concepts as Apple Silicon, in that it combines a CPU, GPU, neural processing elements, and high-speed unified memory on a single chip. Viewed from a high level, the architectures follow the same approach. Evidently, when Apple Silicon stunned the world at its launch, it made an impression on the PC industry. However, despite Nvidia's bluster about its chip being extremely fast and powerful, it does need to be more directly compared against Apple Silicon to see whether it truly stands up. There's no real official benchmark result from Nvidia to compare against at this time, but there was a pre-release benchmark that's noteworthy. Posted to Geekbench in June 2025 and subsequently removed, but archived by Wccftech, the listing for the Nvidia N1x is believed to be an early version of the GPU maker's chip. The version listed includes an ARMv8 chip with 20 cores, a base clock speed of 2.81GHz, and 128GB of unified memory. When it comes to performance, the single-core score is listed at 3,096 points, with the multi-core score reaching 18,837. The immediate comparison made by the publication was to the M3 Max chip in a 16-inch MacBook Pro. On checking Geekbench's listings, the 16-inch MacBook Pro with M3 has a single-core score of 3,128 and a multi-core of 20,969. For reference, the highest M3 Pro result in a MacBook Pro is 3,105 for the single-core score and 15,255 for the multi-core. Under the current M5 generation, the 14-inch MacBook Pro with M5 gets a massive 4,224 for the single-core score and 17,465 for the multi-core. This doesn't seem massively impressive, until you check the core counts of Apple's chips. The M3 Max in question has 16 cores, the M3 Pro has 12 cores, and even the M5 result involved just 10 cores. The current Apple Silicon leader, the 18-core M5 Max, is seen setting scores at around 4,200 again for the single-core, but the multi-core is hovering within touching distance of 30,000. Admittedly the N1x result is a pre-release listing and a year old. It's entirely possible that Nvidia has updated the design, increased the clock speed, and made other changes since that time. However, with modern manufacturing lead times being extremely long, there probably hasn't been much change since then. An impressive first try The main takeaway here is that Nvidia has seen Apple Silicon as a threat, and believes it can do something better for the Windows market. It probably could. Eventually. If the N1x result actually reflects the capabilities of the first chip, it's a good start for the company's initial release. But that said, it's up against some considerable competition. As much as Nvidia boasts about the AI capabilities of RTX Spark, Apple's already got a counter for it. Aside from the Neural Engine, the M5 generation has neural accelerators in each GPU core, making it massively more capable of AI tasks. As it stands, on the CPU front, it is trailing behind a chip from Apple that's more than two years old. Nvidia has some considerable catching up to do.

Nvidia's RTX Spark chip 'reinvents' laptops for agentic AI

Agentic AI requires continuous GPU operation, creating challenges for laptop battery life and portability compared to always-on desktop devices. Nvidia has a vision to put AI productivity into the hands of virtually everyone on the planet. On Sunday night it advanced that mission with the debut of the Nvidia RTX Spark, a challenge to Qualcomm as the second Windows on Arm processor for consumer and business PCs. In conjunction with Mediatek, which designed the N1 and N1X CPU at the heart of the RTX Spark platform, Nvidia is "reinventing the personal computer, for creating, for gaming, for agents," Nvidia chief executive Jensen Huang (or an AI facsimile?) told attendees at Nvidia's GTC conference at the opening of Computex 2026 in Taipei. PC companies, ever quick to jump on a new trend, are on board with Nvidia's vision. Asus, Dell, HP, Lenovo, Microsoft, and MSI showed RTX Spark laptops that will ship in the fall. Acer and Gigabyte will follow that wave. Microsoft, which has pioneered Windows on Arm development with Qualcomm, will release the Surface Laptop Ultra based upon the RTX Spark platform inside. From one perspective, Nvidia's RTX Spark processor is a powerful, yet familiar update to the PC. It's loaded with a 20-core Arm "Grace" CPU, developed by Mediatek, connected to 128GB of unified memory via a 600 GB/s NVLink connection. The kicker is the integrated GPU, which contains a petaflop of AI performance, which Microsoft revealed contains 6,144 Blackwell RTX cores, about the equivalent of an RTX 5070 CPU. We've known about the N1X for some time -- and now it's reached fruition. Will consumers accept Nvidia's framing, which positions that powerful GPU as a gaming device? Huang appeared with two laptops playing Microsoft's Forza franchise and the 007: First Light game. Or will RTX Spark machines be celebrated as AI productivity enablers? Nvidia's PC business is now a fraction of its total revenue, and in fact most of Huang's speech covered the buildout of large data centers using Nvidia's systems and infrastructure. Nonetheless, with RTX Spark Huang showed he hasn't lost touch with the PC line that launched Nvidia decades ago. Huang said AI will be fundamental to the PC experience: "That is the modern application: an [AI] agent." "What becomes of our personal computer in a world of agents, agents running native, connected to models, local or in the cloud, our personal AI sandboxed for security, running continuously, getting work done?" Huang asked. "The chips and the OS must evolve." What we know of RTX Spark PCs Unlike rival CPU makers, which prioritize the speed, core count, number of threads, and other deeply technical characteristics of their CPU cores, Huang downplayed the N1X, barely mentioning it by name. Instead, he focused on the RTX Spark platform itself, and how the RTX Spark platform fit within the overall Nvidia AI ecosystem. An early leak indicated that the N1X would consist of 20 cores, confirmed by Microsoft, made up of Cortex-X925 "extreme " cores and Cortex-A725 cores, with an 18-core variant as well. (In an email, Arm itself confirmed the N1X configuration.) The lower-end N1 could be made from 12 (8+4) cores, as well as a 10 (7+3) configuration. The key will be the number of CUDA cores, which will affect the GPU and AI capabilities: 6,144 or 5,120 in the N1X, and 2,560 or 2,048 CUDA cores within the N1, VideoCardz reported. At the low end, that's about the number in a GeForce RTX 2050. The one metric that we didn't hear: TOPS, which has been a staple of AI hardware discussions for several years, then trailed off. (Microsoft officials confirmed that the RTX Spark platform at the heart of the Surface Laptop Ultra has enough TOPS to qualify it as a Copilot+ PC, but didn't disclose the actual figure.) "RTX Spark lets creators, AI developers and gamers render ultralarge 90GB+ 3D scenes, edit 12K 4:2:2 video, generate 4K AI videos, run 120B-parameter LLMs with up to 1 million tokens context using agents locally, and play AAA games at 1440p and over 100 frames per second," Nvidia said. Still, Nvidia and its partners said that RTX Spark laptops would be available in 14-inch and 16-inch sizes, as thin as 14mm and weighing three pounds on up. Nvidia stated that the laptops would be made from a "precision-machined aluminum chassis" with color-accurate OLED displays. Those laptops, "purpose-built for personal agents," will be complemented by several mini PCs, also with an RTX Spark chip inside, Huang said, showing them off next to the third-party laptops. Huang also showed off several DGX Station desktops, which will be out of reach for most consumers with loadouts of 768GB of memory for running trillion-parameter AI models. For every successive generation of CPU, Nvidia will have a chip for laptops, one for desktops, and one for workstations, he said. Huang said that he envisions the PC evolving much the same way as a smartphone does, which is now barely a phone and more of a pocket computer. "Here's my theory," Huang said. "I could totally imagine that someday there's actually an AI supercomputer in your house, and it's running all of your agents, it's running all of your assistants, and they're doing all kinds of things for you all the time. "You have to have it in your house, just like you have a home theater in your house, you have stereos in your house, you have game consoles in your house," Huang said. "You want assistant AI agent computers running in your house, and these in time becomes a lot more like R2D2 to you. It becomes more like C3PO to you than it feels like a PC to you." Huang said he would have a "conversation" with Microsoft chief executive Satya Nadella talking about their agentic vision for PCs later in CES. Adobe also offered its support, pledging to add a new video pipeline in Adobe Premiere specifically tapping into the RTX Spark unified memory. Adobe's Substance 3D Painter and Stager will run natively on RTX Spark, the company said. Adobe's next-generation Photoshop engine "will be optimized for GPU-accelerated compositing, enabling live filters, high dynamic range and modern natural brushing," it added, with an "AI-native pipeline." Gaming laptops? Sure. But agentic PCs? Color me doubtful What we haven't heard is whether the concept of an agentic PC actually can live in a world ruled by a traditional laptop. By definition, an agentic PC is always active, always connected, sending snippets of code around the Internet to perform the tasks you ask of it. The device that's embodied this vision of productivity so far has been the Apple Mac Mini, a mini PC that's tethered to your desk by at least a power cable, if not an Ethernet cord. Put another way: on the PC, AI depends on the GPU. Agents are always running. By definition, then, so is the GPU. Laptops, by the traditional definition, migrate. They prioritize performance, yes, but also battery life. Gaming laptops with large GPUs typically last a couple of hours at best before running out of juice. I'd like my agents to keep working while I'm on the road...but they can't, not if the laptop is hibernating. Is that what we hope for in the agentic AI era, laptops with constantly spinning fans trapped in a backpack as their owners rush to find a nearby power cable? Or can agents actually run with a laptop closed, quietly getting work done with all-day battery life? Neither Microsoft nor laptop makers have told us. It's very easy to buttonhole the RTX Spark platform as just another gaming laptop, first and foremost. But if these laptops are going to be productivity devices, we'll need to hear how they fit into the paradigm of a productivity laptop. Something's missing. I'm absolutely willing to believe that mini PCs with Nvidia's RTX Spark platform inside can be a success story for AI aficionados working at their desk. I'm much more skeptical about how RTX Spark laptops can succeed as on-the-go, always-connected productivity devices. Nvidia, Microsoft, and hardware vendors will need to convince me that this new platform is in fact better than anything we already have on the market...or just another gaming PC dressed up in a business suit.

Nvidia's RTX Spark Chip to Try and Reinvent the PC for the AI Era

When he's not battling bugs and robots in Helldivers 2, Michael is reporting on AI, satellites, cybersecurity, PCs, and tech policy. After years of rumors, Nvidia is introducing the company's first Arm-based CPU chips for consumer laptops and mini PCs, or what it dubs RTX Spark. But the goal is to go beyond creating a personal computer, and eventually bring AI supercomputers to people's homes. In his Computex keynote in Taiwan, Nvidia's CEO Jensen Huang announced RTX Spark, saying, "40 years later, Microsoft and Nvidia are going to reinvent the PC." The PC chips have not only been designed to be fast and power efficient, but they also promise to run autonomous AI agents, capable of completing tasks for you 24/7, according to Huang. "I could totally imagine some day there is an AI super computer in your house, and it's running all of your agents, it's running all of your assistants, " he added. "And you have to have it in your house, just like you have a home theater in your house." Nvidia plans on launching the first RTX Spark laptops using an "N1X" processor built in partnership with Taiwanese chipmaker MediaTek. The chip uses TSMC's 3 nanometer manufacturing node. The first products will arrive this fall through the top PC makers including Asus, Dell, HP, Lenovo, MSI and Microsoft's own Surface brand. The new product family seems similar to Nvidia's DGX Spark platform, a class of mini PCs that also use the company's CPUs and GPUs, but were designed for AI researchers and developers. The key difference is that RTX Spark is specifically meant for consumers and the Windows 11 OS, whereas DGX Spark runs a custom version of Ubuntu Linux. The RTX Spark "superchip" fuses two "chiplets" together: a GPU based on Nvidia's Blackwell architecture featuring 6,144 CUDA cores, and a 20-core Nvidia Grace CPU. The design appears to be the same as the GB10 superchip in the DGX Spark. Perhaps to stand out from other Arm-based laptops, like those from Apple and Qualcomm, Nvidia's presentation noted the RTX Spark can support up to 128GB LPDDR5X in unified memory, enabling the CPU and GPU to share an extremely large pool of RAM. In return, a user can locally run AI models spanning up to 120 billion parameters, similar to the DGX Spark. The Nvidia laptops have also been built for video and 3D content creation, along with PC gaming. "Microsoft and Nvidia meticulously optimized everything," Huang added, noting the company's Arm-based chip can run any Windows application. Alongside the laptops, RTX Spark will also appear in mini PCs. A slide in Huang's presentation also teased that RTX Spark might expand into Windows-based desktop towers too. The big mystery is pricing and performance benchmarks, along with more specifics about running x86 programs. Nvidia will likely reveal more in the coming months closer to the fall launch. Still, we suspect the RTX Spark is geared more toward power users and AI enthusiasts willing to pay up, considering the up to 128GB in unified memory means the laptops could get very pricey at max specs, especially in light of the ongoing memory shortage. For perspective, Nvidia's DGX Spark features 128GB of RAM, and can be priced from $3,499 to $4,699, depending on the model. Microsoft also told us its own RTX Spark product, the Surface Laptop Ultra, will be the company's most powerful model yet, a sign it won't be cheap. Despite the possible high price, Nvidia's entry into consumer PC chips could shake up the market in other ways. Most notably, it promises to help expand PC gaming to Arm-based processors when x86-based gaming using AMD and Intel silicon has long reigned supreme. Nvidia also appears to be betting that RTX Spark will gain steam when more users have been buying new PCs precisely to run AI agents and models locally, such as OpenClaw. Rival AMD has even been pushing a new product category called the "Agent Computer." Nvidia also showed off a roadmap, indicating the company already has plans for successive generations of RTX Spark chips in store. For customers looking for a tower desktop, the company has started selling the DGX Station, which contains a more powerful Nvidia GB300 chip featuring a staggering 748GB in memory. The pricing for the GB300 has been spotted at over $100,000.

Nvidia and Microsoft tease "a new era of PC" ahead of Computex 2026 -- coordinated social media posts could indicate that rumored N1X laptops will be Windows on Arm systems

An Nvidia-powered Arm PC running Windows could inspire new local AI experiences beyond Copilot+. Ahead of Computex next week, Nvidia's social media accounts have begun promising "a new era of PC," along with the latitude and longitude of the Taipei Music Center, where CEO Jensen Huang will present his keynote for the event as part of GTC Taipei 2026. While we don't have any idea exactly what's coming, it's intriguing to see who else is joining in on the game. The Windows X/Twitter account has shared the exact same message as Nvidia's, suggesting that we could see the long-rumored N1X laptop platform make its debut at Computex - and that it could be running Windows on Arm. For background, N1X has long been rumored to be the mobile variant of the GB10 Superchip at the heart of the DGX Spark mini-PC, which boasts an RTX 5070-class GPU paired with 128GB of LPDDR5X memory and a powerful Mediatek-designed 20-core Arm CPU complex. But the DGX Spark is an Ubuntu Linux-powered AI developer sandbox, not a jack-of-all-trades PC that can seamlessly run Windows apps, as the current crop of Windows on Arm platforms can. If Microsoft is putting its weight behind N1X, that could broaden the appeal of the platform for a more general computing audience by bringing the entire Windows app ecosystem to the platform. Supporting N1X would also bring a powerful, advanced unified-memory-architecture AI computing platform into the Windows camp. None of Microsoft's other Windows on Arm partners have produced anything nearly as ambitious or powerful an AI foundation as the GB10 Superchip, so N1X laptops could be a major boost for the company's AI ambitions on Windows. Having that class of raw compute at its disposal could certainly inspire Microsoft to create new types of first-party local AI experiences that simply haven't been possible from the current crop of Copilot+ PCs and their relatively limited AI grunt. But given what we know about GB10 already, the appeal of this type of system could be narrow at first. Because they share the same pool of LPDDR5X memory, the GB10 GPU enjoys just 273 GB/s of raw bandwidth, far less than that offered by more traditional laptops with dedicated GPUs that have their own pools of GDDR memory. In our own experience, we've found that you can certainly game on GB10, but it's not the platform's strongest suit. So unless there's a major change in the platform's architecture or resources waiting in the wings, N1X PCs will likely need to deliver a new type of experience with their AI potential that's missing from current systems and platform architectures. And N1X PCs will almost certainly be expensive amid the current silicon crunch. GB10 boxes are all selling for around $5000 by our reckoning, and that's partially because they include an exotic NIC that almost certainly won't make its way into any potential laptops powered by this platform. But massive pools of RAM and large SSDs don't come cheap right now, either, so we're still likely to be looking at pricey partner systems. A broader product stack than the 128GB GB10 with smaller memory options and lower CPU and GPU resource counts could help make these systems relatively more affordable while still keeping them plenty powerful for local AI. In short, there's still plenty we don't know about how an N1X-powered AI PC will look, but the fact that Nvidia and Microsoft could be teaming up to make it a Windows on Arm platform is a big deal in itself. We'll be on the ground at Computex 2026 very soon, and we'll report back with details on this potential development as we learn more. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

The PC is being reinvented" - Nvidia announces new AI-focused RTX Spark laptop as widespread doubt bubbles

Nvidia has announced the RTX Spark computer chip, which it claims will revolutionize personal computing thanks to its AI capabilities. This chip, which will be included in PCs and laptops by Dell, Lenuvo, Asus, and HP, will complement Windows software with 1 petaflop of AI performance. It will also feature "full-stack" NVIDIA AI and graphics technology, and up to 128GB of unified memory. In a blog post on the official Nvidia website, CEO Jensen Huang wrote: "The PC is being reinvented. For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask -- and the PC does the work. RTX Spark brings everything NVIDIA has built -- CUDA, RTX, our AI platform -- into a single superchip. Local agents. Frontier models. Creative workflows. RTX games. All on a laptop. This is the new PC. The personal AI computer." The chip will launch in RTX Spark laptops this fall, with the blog post describing the chip as "purpose built for personal agents". As it relates to gaming, the chip will provide the Nvidia suite of Ai and graphics technology to users of the hardware. This includes the raytracing, DLSS, and Reflex technology, which the post states will allow users to play games at 1440p resolution and over 100 frames per second. This announcement comes in the wake of great havoc wrought by generative AI technology in the world of personal computing. The rise of AI companies like Nvidia, and the coinciding demand for hardware used in datacentres, has caused prices for personal computer components to skyrocket. This has also affected console manufacturers. Microsoft and Sony have both raised prices for the Xbox Series X/S and PS5 respectively, and even Valve's Steam Deck hasn't escaped unscathed. Meanwhile, skepticism around the AI industry only continues to rise. While tech companies like Nvidia continue to soar on the stock market due to deep-rooted involvement in the generative AI industry, a 2025 MIT study found the tech may be eroding critical thinking skills. As for the games industry, roughly half of professionals thought it was bad for the industry according to the 2026 state of the industry report.

Not just for AI agents: Nvidia's RTX Spark means Arm-powered laptops for gamers, too, promising 100 fps at 1440p in the latest games

After a lot of leaks, a whole lot of rumours, and me boring the crap out of Nvidia's PRs about it many times over the years, the N1X CPU, now incorporated in the RTX Spark SoC, has finally been unveiled. It's been one of the most well-known secrets in the industry -- that Nvidia would one day release a laptop SoC, pairing a MediaTek Arm chiplet with an Nvidia GPU core, to create its own notebook platform and here we are. With the full RTX Spark "superchip", codenamed the N1X, users can expect up to 20 Grace CPU cores and 6144 RTX Blackwell GPU cores, alongside up to 128 GB of unified LPDDR5x memory. And I think the 'up to' parts of that statement are doing a lot of heavy lifting here, because this isn't just a single laptop processor but a full lineup of notebook SoCs starting in the autumn of this year. And that's welcome news, because in a RAMpocalypse world of ever-increasing hardware prices, a chip that's wedded to 128 GB of any kind of memory is going to be prohibitively expensive. But, while Mark Aevermann, consumer product marketing lead of RTX Spark, wouldn't be drawn on details of the different chips we might expect to see, he did note that "RTX Spark is going to be a family of products that are going to attack a lot of different price points" and that "across the entire family you'll see configurations from 16 GB all the way up to 128 GB." Not that I expect a 16 GB RTX Spark laptop to be cheap but at least that will be a far more affordable SKU, and likely a more gaming-oriented chip, too. Though I wouldn't necessarily expect to see laptops using those lower-end chips at launch in the autumn, as right now Nvidia has only shown off six very premium machines from Asus, Dell, HP, Lenovo, Microsoft, and MSI. "Over time," says Aevermann, "you can expect over 30 laptops and over ten or so desktops initially planned." But with Nvidia promising wattage figures from up to 80 W down to the "low, low single digits" that does make me think about handhelds, especially after Intel's announcement of the G3 Extreme in the past week. Aevermann introduced the RTX Spark as "the most efficient PC chip ever built... a new type of SoC that is both built for the future of computers as agents, as well as a fantastic creation and gaming experience on a single device with exceptional portability." If that doesn't get jammed into a new Shield handheld my ghast will be utterly flabbered. Though Nvidia was also very careful in the briefing ahead of today's announcement to note that it was not making any comments about handhelds. Laptops are not the only form factor Nvidia has been talking about, though, as it's also said that mini PCs, along the same lines as the Linux-powered DGX Spark, will also be releasing in the autumn with the RTX Spark chip inside. The likes of Acer, Asus, Dell, Gigabyte, HP, MSI, and Lenovo will have systems ready to roll, which could make for a real interesting comparison with the Steam Machine whenever that does arrive. The biggest issue for the RTX Spark as a gaming SoC, however, is the fact that it's running a custom Arm and not an x86 CPU core. That means running Windows-on-Arm as the operating system, and that means leaning heavily on the Microsoft's Prism emulator. This is the layer which translates x86 code for WoA devices right now, such as those from Qualcomm, and Nvidia seems pretty confident that it's worked out a lot of the kinks. Though Aevermann does also note that, while you can theoretically get RTX 5070-level performance out of the RTX Spark's GPU core, that is heavily dependent on the application. And I think you can translate that to mean it's completely dependent on how well a given application copes with that emulation layer. Nvidia has a lot of skin in the game here, but also a lot of deep contacts with the gaming world, and it has noted that developers are running the gamut from purely optimising their games for Prism, to porting existing games to Arm, to coding entirely natively for the Arm ecosystem. The company is also engaged with developers to ensure the ecosystem supports all the current anti-cheat software around, something that's been a struggle for Linux gaming. "We are working closely with game developers to ensure all the top games run great," an Nvidia rep tells me. "Our ongoing collaborations are bringing Fortnite, VALORANT, League of Legends, PUBG and more to run on RTX Spark. "One of the biggest challenges is native ARM anti-cheat that online games rely on. We are working with developers to bring support for major anti cheats like Easy Anti‑Cheat, BattlEye, and Denuvo among others. We'll have even more game news to announce over the coming months." Here's hoping Nvidia's confidence in both the Prism emulation and the game developers proves warranted when the RTX Spark systems see the light of day towards the end of the year.

Nvidia launches Windows laptop chip in consumer PC push

Taipei (AFP) - Nvidia unveiled a powerful laptop chip for Windows machines on Monday, staking its claim in the market for next-generation consumer PCs integrated with artificial intelligence. Analysts said the US hardware titan's move challenges the likes of Apple, Intel and AMD in the PC domain, although the new devices will likely carry a hefty price tag. It also represents an attempt by Nvidia -- which the AI boom has made the world's most valuable company -- to diversify into the consumer market, even as it reaps record profits from massive demand for its data centre processors from global tech giants. "Microsoft and Nvidia are going to reinvent the PC. This is going to be the new PC," said Jensen Huang, chief executive of the US tech giant, as he launched RTX Spark ahead of Computex, a major technology show. "If you want to run digital biology, no problem. If you want to do seismic processing, no problem. You want astrophysics, no problem," Huang said. "Microsoft and Nvidia meticulously optimised everything so that this computer literally runs everything the world has ever created, plus it now runs agents, an incredible computer." Nvidia is best known for its GPUs, specialised computer chips originally designed to render video game graphics at high speed, which have more recently become the engine behind AI tools from chatbots to image generators and agents that can carry out tasks for users. As governments and companies pour hundreds of billions of dollars into AI infrastructure, the company's value has topped $5 trillion, more than the gross domestic product of Japan or India. But Monday's announcement instead focuses on a new CPU, or central processing unit, which acts like the brain of a personal computer. "Nvidia is bypassing the traditional PC supply chain to build an end-to-end hardware monopoly," Stephen Wu, a former AI software engineer and founder of the Carthage Capital investment fund, told AFP, calling the news a long-awaited development in the tech industry. Wu called it an "existential threat" to current laptop chip designs, and a strategic attempt by Nvidia to get programmers to build new tech products on their hardware, which will boost data centre GPU demand. "Intel and AMD are the immediate casualties," Wu said, adding that "for AI users, this hardware will finally provide the memory bandwidth necessary to run robust local models without latency". It is not the first time Nvidia chips have powered Windows devices -- a range of tablets did so in the early 2010s. But the new device is positioned as a tool that can easily run AI services such as agents, which have the ability to carry out tasks for users. "This is the first completely re-engineered, reinvented line of PCs that has happened in 40 years," Huang said. "There is no question this reinvention of the computer is as big of a deal as the reinvention of the phone into what we now know as the smartphone."

Nvidia debuts RTX Spark processor for Windows laptops, compact desktops

Nvidia debuts RTX Spark processor for Windows laptops, compact desktops Nvidia Corp. today introduced a system-on-chip designed to power Windows laptops and compact desktops. The RTX Spark, as the processor is called, will roll out alongside Windows upgrades that will make the operating system better at running artificial intelligence agents. Separately, Nvidia will launch an AI-optimized desktop for technical professionals. The RTX Spark is powered by a graphics processing unit based on Nvidia's Blackwell architecture. It features 6,144 CUDA cores, which can perform a wide range of tasks, with an unspecified number of specialized Tensor Cores. The latter circuits are optimized for a narrower set of use cases such as running large language models. Nvidia originally developed Blackwell for data centers. The most advanced server chip that uses the architecture, the Blackwell Ultra, features about four times as many CUDA cores as the RTX Spark. The former chip includes specialized circuits optimized to speed up workloads such as database queries and LLM attention calculations. The RTX Spark combines its Blackwell accelerator with a 20-core central processing unit. The CPU is based on Nvidia's Grace architecture, which was designed for data centers similarly to Blackwell. Nvidia adapted it to Windows computers through a partnership with MediaTek Inc., a major supplier of consumer device chips. The RTX Spark's GPU and CPU exchange data via a custom interconnect called the NVLink-C2C. They're supported by 128 gigabytes of unified memory. Together, the chip's two computing modules provide up to 1 petaflop of performance when processing data in the FP4 format, which is widely used by AI models. An RTX Spark computer can run on-device LLMs with up to 120 billion parameters and a 1 million context window. Such models can be used to power, among other programs, AI agents such as OpenClaw. Nvidia says that the OpenClaw Windows app's developers will integrate the software with a set of new AI agent controls built into RTX Spark computers. Microsoft will extend Windows with "identity, containment, policy and end-to-end security" features for AI agents. Nvidia, in turn, will provide access to its recently launched OpenShell tool. The software runs agents in isolated sandboxes to reduce the impact of cybersecurity incidents. According to Nvidia, the controls will enable users to block agents' access to personal information and limit the range of actions they can perform. "Microsoft and Nvidia are going to reinvent the PC," Nvidia Chief Executive Officer Jensen Huang said at the company's GTC Taipei event today, where it announced the RTX Spark alongside several other product updates. "This is the first completely reengineered, reinvented line of PCs that has happened in 40 years." Running AI agents is not the only use case that Nvidia is targeting with the RTX Spark. The company says that laptops equipped with the chip also lend themselves to running creative applications and video games. Many video games generate visuals using ray tracing, a rendering technique that simulates light beams. The RTX Spark enhances ray tracing programs with a feature called DLSS 4.5 Ray Reconstruction. According to Nvidia, the feature uses a custom transformer model to fix noise in video games frames generated with ray tracing. The company expects DLSS 4.5 Ray Reconstruction to improve lighting and motion efforts. Over in the creative software market, several major players including Adobe Inc. are updating their applications to run on the RTX Spark. Photoshop is receiving enhancements that will enable users to complete some tasks up to twice as fast as before. The image editing application includes several AI-powered features that enable designers to add, remove and change elements with prompts. Nvidia has partnered with more than a half dozen computer manufacturers to bring the RTX Spark to market. Microsoft will also join the fray with the Surface Laptop Ultra, an upcoming high-end laptop. The device combines an RTX Spark with a 15-inch display, five port types and a battery that supports up to 24 hours of use between charges.

Nvidia announces RTX Spark processor for high-end laptops, and it's a huge leap

NVIDIA's new superchip wants to make your laptop your smartest coworker yet. This story is part of our coverage of Computex, the world's biggest computing conference. Updated less than 0 just now ago Nvidia just unveiled the RTX Spark, a new superchip that it says reinvents the Windows PC for the age of personal AI agents. The vision is to transform your laptop from a machine to an actual assistant who can do the work for you. Jensen Huang, founder and CEO of Nvidia, put it in plain terms: "For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask, and the PC does the work." Recommended Videos The RTX Spark combines a Blackwell GPU with 6,144 CUDA cores and a 20-core Grace CPU, connected through Nvidia's NVLink chip-to-chip interconnect. It packs up to 1 petaflop of AI compute and up to 128GB of unified memory, which means it can run massive 120-billion-parameter AI models locally without sending your data to the cloud. What can it do for the new age of Windows laptops? If Nvidia is to be believed, the new RTX Spark processor will usher in a new era for Windows laptops. The chipset can handle extremely demanding tasks, including 90GB+ 3D scene rendering, 12K video editing, and 4K AI video generation. It can also run AAA games at 1440p with over 100 frames per second. For creators, Nvidia is partnering with Adobe to rebuild Photoshop and Premiere from the ground up for RTX Spark, promising up to 2x faster performance across AI editing, color correction, and effects. What makes this different from cloud AI? One of the biggest concerns with running AI on your laptop has always been privacy. Most AI tools today send your queries to the cloud, which means your personal data passes through servers you have no control over. RTX Spark aims to fix that by keeping everything local, on your device, under your control. Nvidia and Microsoft are introducing new Windows security primitives alongside Nvidia OpenShell, a runtime that lets you set strict rules for what agents can and cannot do. It can also mask your personal information before sending any queries to the cloud, so your data stays yours. Vincent Koc, chief architect at the OpenClaw Foundation, is already on board with the approach. "Running solutions like OpenShell and the Microsoft security primitives on RTX Spark will enable users to leverage a fully integrated stack for private, personal agents running on device," he said. If that sounds like something you want on your desk sooner rather than later, you won't have to wait long. RTX Spark laptops will be available this fall from ASUS, Dell, HP, Lenovo, Microsoft Surface, and MSI, with models from Acer and GIGABYTE to follow.

NVIDIA RTX Spark Brings Grace Arm CPUs and Blackwell Graphics to Windows

NVIDIA has officially entered the Windows processor market with the launch of RTX Spark, a new Arm-based computing platform that combines Grace CPU technology with Blackwell graphics. Announced during Computex, the new SoC is designed primarily for AI-focused laptops, though NVIDIA's ambitions extend far beyond mobile systems as the company outlines a multi-generation Windows roadmap. Developed in collaboration with MediaTek, RTX Spark integrates a 20-core Grace processor alongside a Blackwell GPU featuring 6,144 CUDA cores. While that GPU configuration mirrors the core count found on the GeForce RTX 5070 desktop graphics card, the notebook implementation operates within substantially lower power limits, targeting systems with thermal envelopes ranging from roughly 45W to 80W. A key feature of the platform is its unified memory architecture. RTX Spark supports up to 128GB of LPDDR5X memory that is shared between the CPU and GPU. By eliminating separate memory pools, NVIDIA aims to improve efficiency for AI workloads, rendering tasks, content creation, and data-intensive applications. The approach closely aligns with trends seen in modern AI computing, where rapid access to large datasets is becoming increasingly important. NVIDIA positions RTX Spark first and foremost as an AI platform. During its keynote demonstrations, the company showcased local execution of AI agents, coding assistants, large language models, and professional creative workloads. Examples included rendering large 3D scenes using OptiX acceleration, editing ultra-high-resolution video content, and running AI models containing up to 120 billion parameters directly on the device. Gaming capabilities were also highlighted. Demonstrations included titles such as Forza Horizon 6 and 007: First Light running on RTX Spark-powered systems. NVIDIA claims the platform can deliver 1440p gaming with ray tracing enabled when paired with DLSS technologies. While independent performance testing will ultimately determine real-world results, the inclusion of Blackwell graphics gives RTX Spark a considerably stronger gaming profile than most existing Windows-on-Arm solutions. The significance of RTX Spark extends beyond the hardware itself. Until recently, Qualcomm was effectively the sole major supplier of Arm processors for Windows notebooks following its long-standing relationship with Microsoft. With RTX Spark, NVIDIA becomes a direct competitor in the Windows CPU market, challenging not only Qualcomm but also AMD and Intel in selected segments. The project has been years in the making. Industry reports first surfaced in 2023 indicating NVIDIA was developing an Arm-based processor for Windows PCs. Subsequent reports suggested development delays caused by validation issues and software optimization challenges. With the official unveiling now complete, the company appears ready to push aggressively into the Windows ecosystem. Several major PC manufacturers are already preparing systems based on RTX Spark, including ASUS, Dell, Lenovo, MSI, and Microsoft. Initial devices are expected to arrive this fall. NVIDIA also confirmed that RTX Spark represents only the first step in a broader strategy. The company plans to introduce new Windows-focused processors every two years, with future generations expected to incorporate next-generation Rubin graphics technology. Additionally, NVIDIA's DGX Station platform will receive Windows support, further expanding the company's presence in workstation and AI development markets.

Nvidia's N1X could be the jolt Windows laptops need -- with one big catch

However, gaming performance may suffer due to x86 emulation challenges that plague all Arm-based processors, limiting the chip's appeal for gamers. Nvidia is evidently not content to be the world's most valuable company, as the AI and GPU giant now appears primed to dive headfirst into the choppy waters of the laptop processor market. Whether that will help or hurt its fortunes remains to be seen, as the Internet has been aflame this month with rumors that Nvidia will unveil a new "N1X" chip this week at Computex alongside a weaker N1 chip - and the word is both will be SoC (system-on-chip) silicon aimed at Windows laptops. That could be a big deal for anyone who wants to buy a laptop in the next few years, because everything I've heard about the N1X suggests it's optimized for AI performance, battery life, and perhaps even gaming. If Nvidia's efforts to partner with companies like MediaTek and Intel has produced a capable CPU married to a svelte Nvidia GPU on a single chip, utilizing Nvidia's expertise in building high-performance systems for AI and enterprise use, that's potentially a game-changer for the laptop market - and a big challenge to AMD, Apple, and Qualcomm's flagship laptop chips. What is Nvidia N1X? Before I explore where this all could go, let me run down what we know so far. First and foremost, while we don't know for sure if we'll see an N1 chip at Computex, we do know they exist - Nvidia CEO Jensen Huang already confirmed the company has been working on something called the N1 chip for some time. We first got wind of it back in 2024 with reports that MediaTek was considering partnering with Nvidia to make Arm-based chips. At CES the following year Nvidia revealed a new AI-focused DGX Spark mini PC powered by its new GB10 Superchip (basically an Nvidia GPU sandwiched together with a 20-core Arm CPU and a bank of memory, pictured as small box above), and later company chief Huang confirmed (according to Tom's Hardware) that the N1 is basically the same processor that's being used in the DGX Spark and other Nvidia products. So when we started seeing reports (via VideoCardz) early this year that Lenovo was testing laptops with chips sporting N1 and N1X nomenclature, it was pretty easy to guess what was coming. And indeed we've seen reports from the Wall Street Journal that Dell and other manufacturers are working on products packing N1 and N1X chips, suggesting there's a wave of laptops packing Nvidia SoCs coming down the pike. Now, until we see an official press release or fact sheet I can't say for sure how the N1 and N1X chips will stack up against the latest and greatest this year. So far the rumors suggest the N1X could be equipped with a 20-core CPU developed in conjunction with MediaTek, an onboard Blackwell GPU sporting 6,144 CUDA cores (the same amount you find on an Nvidia RTX 5070 GPU) and support for up to 128 GB of LPDDR5X memory in a unified architecture that makes it accessible to both CPU and GPU - just like Apple's M-series chips and Qualcomm's Snapdragon X silicon. Why laptop makers care The fact that Nvidia is reportedly about to launch its own laptop-grade SoC is exciting because Nvidia may be able to find new ways of eking better performance and battery life out of Windows PCs running on these N1 and N1X chips, and it could potentially do a better job of optimizing performance than competitors like AMD and Qualcomm. On the other hand, Nvidia's competitors have all been doing this for years now and they presumably have more data and expertise about what works. We also expect new laptops featuring Apple, AMD, Intel and Qualcomm to hit the market this year, so we'll have to wait and see how the long-rumored N1 and N1X stack up in terms of price and performance. One thing I'm certain of: Nvidia entering the laptop SoC market is going to ramp up competition, and that seems like good news for anyone buying a laptop in the next few years. Why AI may be the real focus Before you start dreaming about thin-and-light MacBook Air competitors packing the power of an Nvidia GeForce RTX 5070 laptop GPU onboard, let me gently bring you back to reality. Remember, Nvidia thinks of itself as more of an AI company than a graphics company at this point, so any new Nvidia-branded laptop chips are almost certainly going to be optimized and marketed for AI applications first and foremost. We've already seen competitors like Qualcomm doing a ton of work to market its Snapdragon X chips as your best choice for AI apps, and indeed the new Snapdragon X2 Elite sports an NPU offering a whopping 80 TOPS (trillion operations per second), outpacing most competitors. Qualcomm's new high-end chip is also a beast when it comes to performance in benchmarks like Cinebench, so if you're passionate about utilizing AI apps or effectively editing video on an ultraportable, a high-end Arm-based laptop could be great for you. Why gamers should be cautious But when it comes to gaming, the story isn't so rosy. Like Nvidia's N1X, Qualcomm's Snapdragon chips are built on Arm architecture, which means they have to emulate an x86 layer in order to run x86 apps. That includes basically all PC games from the last few decades, and while you can play many of them on your new Arm laptop today the performance will be hit-or-miss. This is one of the biggest drawbacks to buying an Arm-based laptop, and by embracing Arm for its new N1 and N1X chips Nvidia is potentially walking right into the same pitfall. In fact the pit is probably deeper, because one of the latest ways to play x86 games on an Arm-based PC is the Prism emulation layer in Windows - and Prism is specifically tuned for Qualcomm's chips, with some performance features that only work on a Snapdragon SoC. So as the Internet salivates over the potential for Nvidia's new chips to power the next generation of gaming laptops, keep in mind that unless they have a solution for the x86 emulation problem the first wave of N1X-powered laptops likely won't blow your gaming laptop out of the water. Nvidia N1X doesn't need gaming to succeed But that's okay, because I don't think Nvidia's N1 chips need to be great for gaming in order to make a difference in how we buy laptops for the foreseeable future. See, Nvidia is getting into the laptop chip biz at a moment when things look especially dreary for high-end notebooks. Between skyrocketing costs for RAM and storage and the already painfully high prices on high-end GPUs, anyone shopping for a powerful PC in the next few years needs either deep pockets, a lot of time to hunt for deals or a willingness to compromise. And if you (like me) are in that last camp, you're prepping for an era in which we focus on scrounging for great deals and buying/building "good enough" hardware rather than chasing the cutting edge. That makes right now an ideal time for something like the MacBook Neo, a perfectly capable (and eye-catching) laptop that can be yours for as low as $600. Cheap Windows laptops are in a really good place right now too, and I expect they'll get even more capable thanks to performance improvements in affordable Intel Wildcat Lake and Qualcomm Snapdragon C laptops debuting this year. If I'm right and we're entering our cheap laptop era, then Nvidia's new N1 and N1X chips could arrive right when we need them most. It would be great to see these new slices of silicon unveiled at Computex 2026 this week with some sort of novel new design or feature that promises to revolutionize the gaming laptop landscape, but I'll settle for stiff competition and an affordable price tag. After all, the last SoC that Nvidia shipped was the Tegra chipset, which is perhaps most famous for driving the Nintendo Switch and Switch 2 consoles - and as good as those machines are, they're not exactly high-performance gaming powerhouses. We'll have to wait until Nvidia's Computex keynote Sunday evening (or Monday morning if you're on site in Taipei) for our best chance to get the scoop on the N1 and N1X chipsets. Stay tuned!

Nvidia's 'RTX Spark' Chip To Try and Reinvent The PC With AI

When he's not battling bugs and robots in Helldivers 2, Michael is reporting on AI, satellites, cybersecurity, PCs, and tech policy. After years of rumors, Nvidia is introducing the company's first Arm-based CPU chips for consumer laptops and mini PCs, or what it dubs RTX Spark. But the goal is to go beyond creating a personal computer, and bring AI supercomputers to people's homes. In his Computex keynote in Taiwan, Nvidia's CEO Jensen Huang announced RTX Spark, saying, "40 years later, Microsoft and Nvidia are going to reinvent the PC." The PC chips have not only been designed to be fast and power efficient, but they also promise to run autonomous AI agents, capable of running tasks for you 24/7, according to Huang. "I could totally imagine some day there is an AI super computer in your house, and it's running all of your agents, it's running all of your assistants, " he added. "And you have to have it in your house, just like you have a home theater in your house." Nvidia plans on launching the first RTX Spark laptops using an "N1X" processor built in partnership with Taiwanese chipmaker MediaTek. The chip uses TSMC's 3 nanometer manufacturing node. The first products will arrive this fall through the top PC makers including Asus, Dell, HP, Lenovo, MSI and Microsoft's own Surface brand. The new product family seems similar to Nvidia's DGX Spark platform, a class of mini PCs that also use the company's CPUs and GPUs, but were designed for AI researchers and developers. The key difference is that RTX Spark is specifically meant for consumers and the Windows 11 OS, whereas DGX Spark runs a custom version of Ubuntu Linux. The RTX Spark "superchip" fuses two "chiplets" together: a GPU based on Nvidia's Blackwell architecture featuring 6,144 CUDA cores, and a 20-core Nvidia Grace CPU. The design appears to be the same as the GB10 superchip in the DGX Spark. Perhaps to stand out from other Arm-based laptops, like those from Apple and Qualcomm, Nvidia's presentation noted the RTX Spark can support up to 128GB LPDDR5X in unified memory, enabling the CPU and GPU to share an extremely large pool of RAM. In return, a user can locally run AI models spanning up to 120 billion parameters, similar to the DGX Spark. The Nvidia laptops have also been built for video and 3D content creation, along with PC gaming. That said, the big mystery is pricing and performance benchmarks. Nvidia will likely reveal more in the coming months closer to the fall launch. Still, we suspect the RTX Spark is geared more toward power users and AI enthusiasts willing to pay up, considering the up to 128GB in unified memory means the laptops could get very pricey at max specs, especially in light of the ongoing memory shortage. For perspective, Nvidia's DGX Spark features 128GB of RAM, and can be priced from $3,499 to $4,699, depending on the model. Microsoft also told us its own RTX Spark product, the Surface Laptop Ultra, will be the company's most powerful model yet, a sign it won't be cheap. Despite the possible high price, Nvidia's entry into consumer PC chips could shake up the market in other ways. Most notably, it promises to help expand PC gaming to Arm-based processors when x86-based gaming using AMD and Intel silicon has long reigned supreme. Nvidia also appears to be betting that RTX Spark will gain steam when more users have been buying new PCs precisely to run AI agents and models locally, such as OpenClaw. Rival AMD has even been pushing a new product category called the "Agent Computer." Nvidia also showed off a roadmap, indicating the company already has plans for successive generations of RTX Spark chips in store. For customers looking for a tower desktop, the company has started selling the DGX Station, which contains a more powerful Nvidia GB300 chip featuring a staggering 748GB in memory. The pricing for the GB300 has been spotted at over $100,000.

Nvidia unveils RTX Spark superchip to power next‑gen AI PCs

NVIDIA has unveiled the RTX Spark, an AI "superchip" designed to enhance AI and graphics performance in Windows laptops and small desktops. The chip boasts 1 petaflop of AI computing power, featuring 6,144 Blackwell RTX cores and 20 Mediatek Arm CPU cores. The RTX Spark is set to compete with AMD's Ryzen AI Max and Qualcomm's Snapdragon X2 chips. NVIDIA claims the performance of the RTX Spark is comparable to the RTX 5070 laptop GPU, while offering significantly lower power consumption. The chip integrates a fast neural processing unit (NPU), which meets the 40 TOPS requirement for Microsoft's Copilot+ initiative. Utilizing a range of unified memory between 16GB and 128GB, the RTX Spark can draw power ranging from single-digit watts to 80W. NVIDIA CEO Jensen Huang characterized the RTX Spark as a major shift in personal computing, suggesting it transforms PCs into devices tailored for AI operations over traditional user inputs. "Today, when you think about your phone, the one thing you don't do with it is make phone calls," said Huang during his keynote at Computex. NVIDIA has collaborated with Microsoft for several years on the design of the RTX Spark, ensuring compatibility with the Prism emulation layer for older Windows applications on Arm-based systems. Additionally, the company is working with leading anti-cheat providers to ensure compatibility for popular games, which has been a barrier for previous systems running on Copilot+. Microsoft has optimized Windows 11's workload profile scheduling specifically for the RTX Spark, according to Pavan Davuluri, head of Windows and devices. "The Windows scheduler on RTX Spark will ensure you get the best performance and efficiency out of your CPU," he stated. The RTX Spark targets content creators and newcomers to AI development, distinguishing itself from the more specialized DGX Spark, which is aimed at large enterprises. NVIDIA's return to launching its own system-on-a-chip products for consumers marks a significant moment since its Tegra line products. The RTX Spark is expected to feature in upcoming devices, including the Surface Laptop Ultra and Dell XPS 16, as well as offerings from various major OEMs.

Lenovo accidentally confirms it is working on laptops powered by NVIDIA's yet-to-be-announced N1X chip

There are plenty of signals pointing toward NVIDIA's rumored N1X chip. The latest comes from Lenovo, whose internal ADFS authentication system has been spotted referencing an "NVIDIA N1x Portal," confirming the OEM is actively working on N1X-powered laptops ahead of Computex 2026. The discovery, spotted by VideoCardz, shows two entries in Lenovo's public sign-in system: "NVIDIA N1x Portal PROD" and "NVIDIA N1x Portal Test." These labels suggest production and test environments for an internal Lenovo portal, confirming that N1X hardware is moving through Lenovo's internal systems. No product specs, model names, or launch dates were revealed. Interestingly, this isn't the first time N1X has broken cover. Earlier support page leaks listed several unreleased Lenovo systems with N1 and N1X labels, including the Legion 7 15N1X11, suggesting a Legion 7 gaming laptop built around the N1X chip. Yoga Pro 7, IdeaPad Slim 5, and Yoga 9 2-in-1 models were also listed, suggesting Lenovo is preparing a wide range of N1X-powered devices across different categories. We also recently covered a laptop motherboard with an N1 chip and 128GB of memory that appeared on Goofish. For those out of the loop, NVIDIA N1X is NVIDIA's upcoming ARM-based chip. Leaks suggest it combines a 20-core CPU and a Blackwell GPU in a single package. The CPU uses a hybrid design with 10 performance and 10 efficiency cores, while the GPU packs 6,144 CUDA cores, the same core count as the desktop RTX 5070. The chip is built on a 3nm process and supports up to 128GB of LPDDR5X memory. The N1X is likely based on the same GB10 Superchip found in NVIDIA's DGX Spark, with the laptop version expected to ship at a lower power target. NVIDIA CEO Jensen Huang confirmed earlier this year that the company is designing the chip in partnership with MediaTek, describing it as offering "low power consumption but excellent performance." If the specs hold up, this would mark the first time a Windows ARM laptop could realistically handle gaming, video editing, and AI workloads without a discrete GPU. NVIDIA is expected to reveal the N1X at its Computex 2026 keynote. If it doesn't surface there, it is hard to say when it would, given that further delays would put it in direct competition with Qualcomm's Snapdragon X2 Elite Extreme and Apple's M5 Pro at an increasingly disadvantageous time.

These new NVIDIA laptops finally give Windows what it's been missing: real creative power

* RTX Spark announced: an AI-forward ARM superchip with a 20-core Grace CPU and Blackwell GPU that delivers up to 1 petaflop and 6,144 CUDA cores in thin laptops. * Native Adobe apps offer hiccup-free video editing, including 90GB 3D scenes and 12K HDR video. * First laptops with this chip will arrive this fall from Microsoft Surface, Dell, Asus, MSI, and more. NVIDIA and Microsoft have partnered to create the RTX Spark, a new, lightweight yet powerful chipset that's shaking up the power structure once again for next-gen computers. It's an ARM-based superchip designed to handle heavy video and 3D workloads directly on Windows PCs, and is even said to hold its own for AAA gaming. Like the CoPilot Plus CPUs it's competing with, such as Qualcomm's Snapdragon Elite X2, the platform targets high-end laptops and compact desktops with great power efficiency. It's also said to offer enough headroom for agentic-forward AI platforms, offering enough local computing power to bypass the cloud entirely. Qualcomm's Snapdragon C platform could shake up the budget PC landscape This looks like a huge move for Arm PCs Posts By Dave Schafer How RTX Spark supercharges content creation and gaming One area most Windows laptops still struggle The architecture addresses the exact bottlenecks that have historically slowed down digital creators, an area where Windows laptops without dedicated GPUs still struggle. At its core is a custom MediaTek-designed 20-core Grace CPU paired with a Blackwell RTX GPU. That combination yields up to 1 petaflop of processing power, aided by 6,144 CUDA cores and up to 128GB of unified memory. The beauty is that you'll get all that power in laptops as thin as 14mm and weighing around 3 pounds. Adobe is already rebuilding Creative Cloud apps like Premiere and Photoshop to run natively on Spark. The direct integration is expected to double the processing speed of automated editing tools. For more traditional workflows, the hardware has enough raw power to render 90GB 3D scenes and edit 12K HDR video formats without stalling. Gaming also figures to be dreamy on these devices, with NVIDIA teasing an environment where AAA games can run at resolutions up to 1440p and 100-plus frames per second with ray tracing and DLSS (up to 4.5 for now). RTX Spark laptops could be the best for local AI NVIDIA wants to enable AI everywhere you can tote a notebook Raw rendering power is just part of the pitch. The RTX Spark shifts how Windows handles automated workflows by moving the processing away from external servers. Microsoft and NVIDIA are deploying new security primitives alongside the NVIDIA OpenShell runtime to manage permissions. This setup routes tasks to local models and masks any personal information before it leaves the device. The system also has ridiculous compute headroom: it apparently can run 120-billion-parameter models natively, keeping data entirely on the machine. Subscribe for deep takes on RTX Spark and PC AI Unlock in-depth newsletter coverage by subscribing -- clear, hands-on analysis of RTX Spark, PC AI, and next-gen chipset impacts for creators and gamers on Windows. Subscribe for expert breakdowns and practical performance context. Get Updates By subscribing, you agree to receive newsletter and marketing emails, and accept our Terms of Use and Privacy Policy. You can unsubscribe anytime. Spark is said to handle 1 million token context windows without breaking a sweat, putting it in leagues with cloud-based offerings such as Google Gemini and Anthropic's Claude. These could turn out to be the best all-in-one devices for running agentic AI models. You can be sure Microsoft's Surface line will be among the first to wield this new power, and Asus, Dell, HP, Lenovo, and MSI are also confirmed to be making new devices. The first wave of RTX Spark laptops arrives this fall.

Nvidia's N1X processor for laptops could be right around the corner

An Nvidia ARM gaming chip in a Legion laptop? Things are getting interesting. Lenovo has accidentally confirmed it is working on laptops powered by Nvidia's yet-to-be-announced N1X chip. The confirmation comes from Lenovo's own ADFS authentication system, which referenced an "Nvidia N1x Portal" in its internal login page, as first spotted by the folks at VideoCardz. Earlier support page leaks also listed several unreleased Lenovo systems with N1 and N1X labels, including the Legion 7 15N1X11, which points to a Legion 7 gaming laptop built around the N1X chip. So what exactly is the Nvidia N1X? According to leaks, the N1X is Nvidia's upcoming ARM-based chip that combines a 20-core CPU with a Blackwell GPU in a single package. The CPU uses a hybrid design with 10 performance cores and 10 efficiency cores, and the GPU packs 6,144 CUDA cores, which is the same core count as the desktop RTX 5070. The chip is built on a 3nm process and supports up to 128GB of LPDDR5X memory. The N1X is probably the same chip powering Nvidia's DGX Spark compact AI computer, which runs at 120W. The laptop version will likely ship at a lower power target, which means slightly dialed-back performance, but still a massive leap over anything currently available in a Windows ARM laptop. Could this change Windows gaming laptops? If the chip turns out to be as good as it seems on paper, for the first time, a Windows ARM laptop could realistically handle gaming, video editing, and AI workloads without needing a separate graphics card. The big caveat is software. Windows on ARM has improved a lot, but game compatibility and driver support are still works in progress. If Nvidia can sort that out, the Legion 7 N1X could be a genuinely exciting laptop. Recommended Videos I have been waiting for a Windows on ARM laptop that can genuinely compete with Apple's MacBook Pros, and it seems the upcoming Nvidia N1X-powered laptop will finally deliver it.

Your next laptop could have an Nvidia CPU - here's why that's a huge deal

Nvidia RTX Spark silicon will rival Intel, AMD and Qualcomm - and the Microsoft Surface is first in line Have we just entered "a new era of PC"? That was Nvidia's bold claim as it officially revealed the RTX Spark, a new processor aimed squarely at the Intel/AMD (and to a lesser extent, Qualcomm) establishment. Big-name manufacturers including Dell and Microsoft are already queuing up. Here's what it could mean for your next laptop. The RTX Spark is essentially a Windows-friendly version of Nvidia's GB10 'superchip', which powers the firm's AI- and developer-focused DBX Spark mini-PC. Each one has multi-core CPUs based on ARM architecture and graphics cores based on the same Blackwell platform as the firm's 5000 series graphics cards - which should mean they'll be potent performers for desktop duties, creative apps and games, as well as on-device artificial intelligence. Nvidia says it has been working with Microsoft to get Windows 11 in better shape to run AI agents on-device using the new hardware. Early leaks suggested there would be multiple versions the chip, codenamed N1 and N1X, but so far Nvidia is only talking about the top-end RTX Spark. It will have 20 CPU cores and 6144 CUDA cores, support up to 128GB of unified memory, and run in a 45-80W power window. Nvidia will likely be targeting AMD's 'Strix Halo' Ryzen processors and Intel's Series 3 Core Ultra 9 chips here. Performance looks promising on paper, with Nvidia touting 12K video editing, 100fps AAA gaming at 1440p resolution, and the ability to run 120 billion-parameter local language models (LLMs) on-device. "All-day" battery life is also on the cards. A step-down model is expected to get 18 CPU cores and 5120 CUDA cores, while more mainstream versions will reportedly have either 12 CPU cores and 2560 CUDA cores, or 10 cores and 2048 CUDA cores. Maximum memory capacity here should be 64GB, and power range is a lower 18W to 45W. That'll make them better suited to lightweight laptops, with Intel's Core 7, AMD's Ryzen 400 and Qualcomm's Snapdragon X2 Elite being the competition. There'll be RTX Spark laptops with 14in and 16in screen sizes; all promise aluminium chassis, OLED display panels with Nvidia G-sync variable refresh tech. Among the first will be a version of the Dell XPS 13; an Intel variant of the (surprisingly affordable) ultraportable was only revealed earlier this week. Asus and Lenovo will reveal their initial RT Spark efforts imminently. Microsoft is also gearing up to launch an all-new Surface. This won't be the first time Nvidia silicon has powered a Surface device. Back in 2012 the Surface RT used an Nvidia Tegra chip and ran the cut-down Windows RT 8.1. Microsoft has learned plenty about running Windows on non-x86 hardware in the decade and a half since then, although a lot of recent Windows on ARM groundwork was done by Qualcomm for its Snapdragon X and X2 Elite chips. We can expect laptops to go on sale around September 2026. The biggest question right now is pricing. Announced during the firm's keynote presentation at the Computex trade show in Taiwan (home turf for CEO Jensen Huang - you'll see his face countless times in photos pinned to Taipei's night market stalls), RTX Spark could potentially be the biggest shakeup of the PC industry in decades. While Qualcomm made a splash with Snapdragon X Elite in 2025 - and stands to gain even more ground with the more powerful Snapdragon X2 Elite this year - it's still very much in third place behind front runners Intel and AMD. If Team Green wants a piece of that action, it's hard to see what those rival firms can do about it: Nvidia is the world's most valuable company, with an outrageous market value of over $5 trillion. Almost all of that value came in the last six years, thanks mainly to the firm's starring role in the booming AI industry. The tech world's obsession with artificial intelligence has also sent the cost of memory and SSD storage sky-high, leading to dramatically increased prices for PC hardware and consumer electronics in general. With Nvidia also using its Computex keynote to big up its AI supercomputers, that's unlikely to change any time soon.

NVIDIA and Microsoft bet on AI agents to reinvent the personal computer

NVIDIA and Microsoft are set to launch new Windows PCs powered by AI. These devices will act as intelligent assistants, performing tasks for users. This technology aims to bring advanced AI capabilities directly to personal computers. Manufacturers like ASUS, Dell, and HP will offer these new AI-focused machines later this year. NVIDIA and Microsoft are making a case for what could be the next major shift in personal computing: PCs designed not just to run applications, but to act as AI-powered assistants capable of carrying out tasks on behalf of users. At NVIDIA's GTC Taipei conference, the company unveiled RTX Spark, a new AI-focused computing platform that will power a new generation of Windows laptops and desktops from manufacturers including ASUS, Dell, HP, Lenovo, Microsoft Surface and MSI. The announcement comes as the technology industry increasingly moves beyond chatbots toward AI agents, software systems capable of reasoning, planning and executing multi-step tasks. While cloud-based AI services have dominated the conversation so far, NVIDIA and Microsoft are now pushing for a future where many of these capabilities run directly on users' devices. "With RTX Spark and Microsoft Windows, you ask and the PC does the work," NVIDIA CEO Jensen Huang said while unveiling the platform. A PC Built Around AIAt the heart of RTX Spark is a new superchip that combines NVIDIA's Blackwell graphics architecture with a custom Arm-based CPU developed in collaboration with MediaTek. NVIDIA claims the platform can deliver up to one petaflop of AI computing performance while supporting as much as 128GB of unified memory, specifications aimed at handling increasingly large AI models locally. The company says users will be able to run large language models with up to 120 billion parameters, process context windows reaching one million tokens, generate AI video, edit ultra-high-resolution content and execute complex AI workflows without relying entirely on cloud infrastructure. The hardware is also designed to retain traditional PC capabilities. NVIDIA says systems powered by RTX Spark will support AAA gaming at over 100 frames per second at 1440p resolution while also handling professional creative workloads. The Push for On-Device AgentsPerhaps the more significant announcement was not the hardware itself, but the software framework NVIDIA and Microsoft are building around it. The two companies are introducing a new Windows-native environment for AI agents, complete with security controls designed to govern what agents can access and how they interact with user data. A new runtime called OpenShell will allow users to define boundaries for AI agents, determine what information can be shared externally and route sensitive requests to local AI models instead of cloud services. The move addresses one of the biggest challenges facing agent-based computing today: trust. While AI agents are becoming increasingly capable, many users remain reluctant to grant them broad access to files, applications and personal information. NVIDIA and Microsoft are betting that running these systems locally, with stronger controls, could accelerate adoption. Microsoft CEO Satya Nadella described the initiative as part of the company's broader ambition to bring what he called "unmetered intelligence" to homes and workplaces. Adobe and Other Developers Join InThe ecosystem around RTX Spark is already beginning to take shape. Adobe said it is redesigning key components of Photoshop and Premiere Pro to take advantage of the new platform, with the goal of delivering significantly faster AI-assisted editing and content creation workflows. According to NVIDIA, creative applications will gain direct access to the platform's unified memory architecture and AI acceleration capabilities, enabling faster rendering, editing and generative AI features. Beyond Adobe, more than 100 software developers and gaming companies are supporting the platform, including Blackmagic Design, Blender, CapCut, ComfyUI, NetEase, Remedy Entertainment and Xbox. For developers working with AI models, NVIDIA is positioning RTX Spark as a portable workstation capable of running advanced local AI workloads that previously required desktop-class hardware or cloud infrastructure. A New Category of Windows PCsThe first RTX Spark-powered devices are expected to arrive later this year. Manufacturers are preparing a range of products, from ultra-thin laptops aimed at creators and developers to compact desktop systems designed for AI workloads and gaming. The larger ambition, however, goes beyond a hardware refresh cycle. For decades, personal computers have largely operated through applications launched and controlled directly by users. NVIDIA and Microsoft are proposing a different model, one where software agents become intermediaries between users and their devices. Whether consumers embrace that shift remains to be seen. But as AI increasingly moves from answering questions to performing actions, NVIDIA's latest announcement suggests the PC industry believes the next computing platform may not be defined by apps, but by agents.

NVIDIA's RTX Spark Is a Direct Shot at the PC Market, Backed by a Multi-Gen Roadmap That Past 'Windows on Arm' Bids Never Had

NVIDIA has announced RTX Spark, its brand-new PC platform, designed in collaboration with Microsoft, as it starts a new chapter of its client brand. NVIDIA RTX Spark May Look Like The Standard AI SoC Affair For PCs, But It's Much Bigger Than That Today at GTC Taipei 2026, a conference hosted by NVIDIA and presented by CEO Jensen Huang, the company unveiled its brand new chip for the PC segment, which they are calling RTX Spark. According to NVIDIA, the RTX Spark reinvents the PC. This project was initiated three years ago with Microsoft, Arm, MediaTek, and other ecosystem partners. Last year, NVIDIA introduced the DGX Spark. The DGX Spark was designed as an AI-first machine, with 128 GB of LPDDR5X memory, the Grace CPU core, and Blackwell GPU architecture. It's like NVIDIA taking its Grace Blackwell datacenter GPUs and optimizing a configuration that is suitable for small and more accessible form factors. The RTX Spark builds upon DGX Spark's GB10 Superchip, which is also internally referred to as N1X, with a lighter configuration called N1. These SoCs are called DGX Spark for the AI, & RTX Spark for the client platforms. RTX Spark Specs Deep-Dive So let's start with the specifications first. NVIDIA's RTX Spark SoC is very much a GB10 Superchip that combines the two aforementioned architectures through Connect-X. NVIDIA developed the GB10 Superchip, which combines innovations from datacenters, such as NVFP4, CUDA, SLANG, TensorRT, vLLM, CX-7 NIC, NVLINK C2C, TMEM, and more, down to client platforms that utilize a smaller form factor, made possible using multi-die packaging tech, a very low-power C2C interface, and Unified Memory Architecture (UMA). As a result, the DGX Spark Workstation was built, which offers the following key features and benefits: * GB10 Grace Blackwell Superchip: Accelerates AI, Data Science, Compute, Rendering & Visualization * 128GB Coherent Unified System Memory: Works with Large AI models up to 200 billion parameters, fine-tune models of up to 70 billion parameters * ConnectX-7 Networking: Connect two DGX Spark systems to work with models of up to 405b parameters * DGX Base OS and NVIDIA AI Software Stack: Seamlessly move workloads from DGX Spark to DGX Cloud or any accelerated data center or cloud infrastructure * Flexible deployment configurations: Configure as an AI Workstation or a network-connected personal AI cloud * Great Desktop Experience: Multi-head display support and flexible connectivity * Compact, power-efficient design: Easily fits on any desk, powered by a standard wall outlet So let's dive into the specifications of the GB10 Superchip. First up, we have the SoC composition, which shows that the chip itself is composed of two dielets, an S-Dielet which houses the CPU, memory subsystem, etc, and a G-Dielet which houses the GPU core. These two dielets are packaged together using Advanced 2.5D packaging and are fabricated on TSMC's 3nm process technology. The CPU is based on the ARM Arch v9.2 architecture with 20 cores, custom-designed by MediaTek. There are 2 clusters of 10 cores each, and each core has a private L2 cache and a 16 MB L3 cache per cluster, so 32 MB in total. The GPU is based on the GB100 Blackwell architecture and is considered an iGPU since it is on the same package and silicon. It features 5th Gen Tensor Cores with DLSS, Reflex, ACE, G-Sync, REFLEX, CUDA, NVFP4 support, and RTX Ray Tracing cores. It produces up to 31 TFLOPs of FP32 and 1000 TOPS of NVFP4 (FP4) compute for AI workloads. The GPU also gets an additional 24 MB of L2 cache. Moving into the memory system, the NVIDIA GB10 Superchip SOC features support for 256b LPDDR5x (UMA) with up to 9400 MT/s speeds, enabling up to 301 GB/s of raw bandwidth, and up to 128 GB of maximum capacities. The system fabric is a high-performance coherent fabric that offers support for CHI-E Coherency Protocol. The GPU has access to the entire system bandwidth of 600 GB/s (Aggregate) over the C2X interface. NVIDIA will allow the unified memory pool to be allocated manually to the GPU. A 128 GB system can dedicate up to 111 GB of system memory to the GPU without opening the UEFI BIOS; it will be done directly through Windows. There's also 16 MB of System Level Cache, which serves as L4 for the CPU, and enables power-efficient data-sharing between the multiple engines on the SoC. The C2C interface is also high-bandwidth and low-power, enabled through NVIDIA's NVLINK architecture. On the connectivity side, NVIDIA's GB10 Superchip SoC offers PCIe, USB, Ethernet over PCIe, and drives up to 4 concurrent displays (3 DP + 1 HDMI) at up to 4K @120Hz with DP Alt-mode, and up to 8K @ 120Hz with HDMI 2.1a. Security features include Dual Secure Root support, SROOT processor, OSROOT processor, and support for both fTPM and discrete TPM. The whole chip has a TDP of 140W. Following is the block diagram of the NVIDIA GB10 Superchip SoC: Scalability is also another fun aspect of the GB10 Superchip. You can connect multiple GB10 chips through NVIDIA's ConnectX Technology and scale throughput, bandwidth, and DRAM capacities to support larger AI models. The ConnectX NIC is connected to the GB10 SoC using a PCIe Gen5 x8 interface, and the units communicate with each other using Ethernet. The Software Enablement & Early Previews Why I say that NVIDIA has a better chance than others in crunching away client PC market share from x86 giants such as Intel and AMD is because, unlike Snapdragon X1 and X2 series, which were a first attempt by Qualcomm on the PC, NVIDIA has a full-fledged and optimized framework around Windows. The answer is simply: RTX. NVIDIA's RTX platforms have already been powering desktops and laptops. The company has a very solid driver support, and its Linux support has also improved substantially, with DGX Spark systems seeing some big updates. TensorRT and CUDA support remain great on Windows, & that gives NVIDIA a better chance at the "WoA" segment than what others have tried in the past. One of the biggest challenges for NVIDIA won't be the GPU, but the CPU. Grace currently remains entirely untested as a Windows chip. The benchmarks that have been floating around for more than a year are all DGX-based through emulation or on Linux. During our technical sessions with the GeForce Team, we saw games and applications such as Blender, Unreal Engine, LLM Studio, OpenClaw, and many more running at reasonably fast performance. In closed testing done by NVIDIA, the chip was getting up to 10x faster AI performance, and up to 2x faster GenAI performance. The gaming performance in titles such as Alan Wake II, Fortnite, and Indiana Jones was great too. But the numbers have to be taken with a grain of salt, and NVIDIA isn't providing any FPS figures, nor showing the settings used to run these games. Though we were told that the games were running with DLSS Frame-Gen enabled. RTX Spark, being based on the Blackwell IP, means that it will support all innovations such as DLSS 4.5 Ray Reconstruction, Multi-Frame Gen, and much more. NVIDIA is also working with various partners, including the XBOX Team and Krafton, to bring next-gen experiences to RTX Spark. A Multi-Generational Plan The key to success for any platform is to make a statement that it won't just end after one launch. For RTX Spark, NVIDIA fully commits to the PC platform, & offered a look into what comes after the first-gen RTX Spark systems. In the first generation, NVIDIA RTX Spark systems will combine the Grace CPU & Blackwell GPU with LP5X memory. These will be available in the fall of 2026. The next chapter of RTX Spark will arrive by 2028, offering Vera CPU and Rubin GPU architectures with LPDDR6 memory support. Then in 2030, NVIDIA will update RTX Spark with its next-generation Rosa CPU and Feynman GPU architectures. One thing to note is that each generation of RTX Spark SoCs will have two variants: a higher-end, such as the 1 PFLOPs Grace Blackwell design, and a lower-end, such as the 400 TFLOPs Grace Blackwell design. NVIDIA RTX Spark "PC" Roadmap: First Laptops From NVIDIA Partners The first NVIDIA RTX Spark laptops will launch this fall. NVIDIA has announced a total of six designs so far, which include the following: * MSI Prestige N16 FLIP AI * Dell XPS 16 * Lenovo Yoga Pro 9n * ASUS ProArt P14 * HP OmniBook Ultra 16 * Microsoft Surface Laptop Ultra Now, one question is the pricing and configurations. NVIDIA has mentioned up to 128 GB, and we know that DGX Spark and OEM variants of the machine have been shipping only with 128 GB configs. If that's going to remain the case with RTX Spark, then these laptops are going to be priced way high up in the premium category due to DRAM supply shortages. We hope there are RAM/SSD configurations to select from. But the RTX Spark story doesn't end with laptops; it is a widely accessible solution that will also make its way to desktops in the form of Mini PCs. These will also be available in Fall 2026, & will include designs from Acer, ASUS, Dell, Gigabyte, HP, MSI, and Lenovo. As NVIDIA continues to develop its DGX and RTX Spark ecosystems, we can potentially see proper Desktop PCs utilizing it, filling the gap between the RTX Spark Mini PCs/Laptops and DGX Workstations. Follow Wccftech on Google to get more of our news coverage in your feeds.

Surface Laptop Ultra is the first laptop to feature Nvidia RTX Spark

After days of teasing from both Microsoft and Nvidia, the mystery has finally been solved. A few days ago, both Microsoft and Nvidia began simultaneously hinting on social media that something "big" related to hardware was in the works. That something has now finally been revealed in the form of the Surface Laptop Ultra. A sleek little device that also becomes the first to feature Nvidia's new ARM-based RTX Spark. The Surface Laptop Ultra is equipped with a 15-inch mini-LED display with a resolution of 262 pixels per inch and a peak brightness of a whopping 2,000 nits. Something Microsoft itself describes as the brightest Surface display ever. The laptop will also come equipped with a haptic touchpad (finally!). The RTX Spark is based on Nvidia's Blackwell architecture and can be configured with up to 20 CPU cores, 6,144 GPU cores, and 128 GB of unified memory. It's also claimed that the chip is powerful enough to deliver a petaflop of graphics and AI performance, on par with certain laptops equipped with an RTX 5070. No price has been revealed yet, but given the current market conditions and the performance and build quality that the Surface Laptop Ultra appears to be aiming for, we should probably expect a rather hefty price tag. However, Microsoft isn't the only one experimenting with the new RTX Spark platform, and several other manufacturers are expected to unveil their own laptops this year featuring Nvidia's new chip under the hood, including ASUS, Dell, HP, and MSI.

Nvidia Launches RTX Spark AI Superchip for Windows Laptops and Desktops

Nvidia has officially unveiled the Nvidia RTX Spark, which might mark the beginning of a major transformation for Windows PCs in decades. The RTX Spark is a new AI-focused superchip that is designed to power the next generation of Windows laptops and desktops. The superchip promises to provide up to 1 petaflop of AI performance, 128 GB of unified memory, and full RTX gaming and creative capabilities packed into a slim device. So, gamers, creators, and developers can run massive AI models locally. The amount of workload that traditional PCs would struggle to handle will be easily managed by systems with the Nvidia RTX Spark built into them. Nvidia Announces New AI Superchip RTX Spark Tailored for Windows PCs The Nvidia RTX Spark combines an Nvidia Blackwell RTX GPU featuring 6144 CUDA cores with a custom 20-core ARM-based Grace CPU. Nvidia CEO Jensen Huang states, "The PC is being reinvented." It will transform from a simple click-and-type device that launches apps into something that can work alongside users through AI agents. All you need to do is command, and the PC will do the work for you. Nvidia and Microsoft are collaborating to bring Windows-native AI agents to these systems using the new security tools and Nvidia OpenShell technology. These agents are designed to securely perform tasks across the apps, like searching for local files, assisting with coding, and even automating workflows. The headline that has grabbed the attention of the gaming community is the 128GB of unified memory. This dramatically expands the AI capability because users can run 120-billion-parameter large language models with up to 1 million tokens of context. The best part is that you don't need to rely on the cloud for executing this on your PC. The Nvidia RTX Spark platform is reportedly said to be capable of rendering 90GB + 3D scenes and editing 12K 4:2:2 video. Furthermore, it also states that you will be able to generate 4K AI video and play AAA games at 1440p above 100 FPS with RTX technologies like Nvidia DLSS and Reflex. The gaming support looks strong as Microsoft confirmed Xbox experiences are coming to the Nvidia RTX Spark systems. Moreover, developers like Riot Games, Remedy Entertainment, and NetEase are preparing titles optimized for this platform. The hardware in itself is quite ambitious, as Nvidia RTX Spark laptops will launch in premium 14-inch to 16-inch designs. These are stated to have all-day battery life, be as thin as 14mm, with tandem OLED displays and lightweight aluminium builds. The community reactions have already started pouring in, as they are excited about what this Nvidia superchip promises.

NVIDIA RTX Spark Wants to Turn Your Windows PC Into a Personal AI Assistant

NVIDIA and Microsoft are betting that the next major evolution of the personal computer will be driven by AI agents rather than traditional applications. With the introduction of RTX Spark, the companies are bringing together dedicated AI hardware, Windows-native agent capabilities and NVIDIA's broader software ecosystem in an effort to create a new category of AI-first PCs. * Make Telecom Talk My Trusted Source Announced at NVIDIA GTC Taipei, RTX Spark is a new superchip designed for Windows PCs that combines AI processing, graphics performance and energy efficiency in a single platform. NVIDIA says the technology is designed to help PCs move beyond simply launching applications and towards understanding requests, completing tasks and assisting users through AI-powered agents that can run directly on the device. The announcement reflects a broader shift taking place across the technology industry as companies increasingly look at ways to make AI a more integrated part of everyday computing rather than something accessed solely through cloud services. What Is NVIDIA RTX Spark? At the heart of RTX Spark is a custom architecture that combines an NVIDIA Blackwell RTX GPU featuring 6,144 CUDA cores with a 20-core NVIDIA Grace CPU. The two chips are connected using NVIDIA's NVLink-C2C interconnect technology, allowing them to work together more efficiently.

Nvidia launches chip for Windows laptops in consumer PC push

"Microsoft and Nvidia are going to reinvent the PC. This is going to be the new PC," said Jensen Huang, chief executive of the US tech giant, as he launched RTX Spark ahead of Computex, a major technology show. kaf/amj/ane Nvidia unveiled a powerful laptop chip for Windows machines on Monday, staking its claim in the market for next-generation consumer PCs integrated with artificial intelligence. "Microsoft and Nvidia are going to reinvent the PC. This is going to be the new PC," said Jensen Huang, chief executive of the US tech giant, as he launched RTX Spark ahead of Computex, a major technology show.

MSI Pushes NVIDIA's RTX Spark Into The Mainstream With A Developer Mini PC And A Tandem OLED Flip Laptop

Vendors have been quick to adopt the latest NVIDIA RTX Spark SoC in their devices, and MSI has deployed it in both mini PCs and laptops. MSI Power EdgeMesa N AI+ Mini PC and Prestige Flip AI+ Laptop With NVIDIA RTX Spark, Delivering a Powerful Arm+Blackwell Combo for Strong Compute and AI Performance With the introduction of the "New era of PC" by NVIDIA, its partners are quickly adopting its recently developed RTX Spark SoC, which brings 20 Arm-based CPU cores coupled with 6144 CUDA Cores based on the Blackwell architecture. The SoC is a powerful chip that brings 128 GB of unified LPDDR5X memory and up to 1 PFLOP of AI performance, which has made it possible to bring powerful compute and local LLM inference to mainstream computers. MSI, being one of the major NVIDIA partners, has adopted the RTX Spark to ready its new-generation devices. The latest stack includes a mini PC and a laptop, both of which aim at high-performance workloads, AI-based creative workflows, gaming, etc. The first device is the EdgeMesa N AI+ Mini PC that is designed for developers, data scientists, and creators. Ideal for LLMs and generative AI, the machine has all the horsepower to carry out operations quickly. EdgeMesa N AI+ offers flexible I/O expansion and modern connectivity that includes multiple high-bandwidth ports. The machine can support four displays at a time through a single HDMI port and 3x USB Type-C ports. All these ports are situated at the back alongside a 10 GbE LAN port for fast internet speeds. MSI says the EdgeMesa N AI+ is ideal for industries such as healthcare, retail, finance, robotics, and smart city applications, and even though it would be perfect for consumers, it doesn't appear to be one made for the mainstream market. For consumers, MSI has a different offering called Prestige N16 Flip AI+, a consumer-grade laptop that brings the power of RTX Spark for heavy workloads and gaming. It's a pretty thin and lightweight machine that brings the best of the industry, including a top-notch Tandem OLED display, flexible design for improved productivity, and modern connectivity. The laptop features a 16-inch UHD+ display that runs at up to 1000 nits of peak brightness and features 100% DCI-P3 coverage for vibrant visuals. It's perfect for both gaming and content creation, thanks to the powerful hardware that sits inside and the color-accurate display that brings Delta E < 1 color accuracy. MSI hasn't shared the pricing and availability details of both machines yet, but these will cost thousands of dollars for sure. News Sources: MSI, MSI Follow Wccftech on Google to get more of our news coverage in your feeds.

Microsoft and NVIDIA unveil next-gen Windows PCs powered by RTX Spark Superchip

At the NVIDIA GTC conference, Microsoft and NVIDIA announced a multi-year, full-stack collaboration to deliver a new class of thin-and-light Windows PCs. Accelerated by the newly unveiled NVIDIA RTX Spark™ superchip, these systems are engineered specifically for developers, creators, and power users, with a core focus on executing localized AI agent workflows. This platform integrates Windows operating system optimizations with NVIDIA's silicon and graphics architecture to shift personal computing from a tool-based model to an interactive, agent-driven teammate experience. Technical Architecture and Deep OS Optimization The RTX Spark superchip features up to 6,144 Blackwell architecture CUDA cores, fifth-generation Tensor Cores with FP4 precision, and a 20-core CPU built on the Arm architecture via a design partnership with MediaTek. Equipped with up to 128GB of unified memory and delivering 1 petaflop of AI performance, the platform relies on tailored Windows 11 architectural updates to maximize its heterogeneous hardware. Performance and Power Management * Workload Profile Scheduling (WPS): Microsoft implemented and optimized WPS specifically for RTX Spark, allowing the Windows scheduler to dynamically distribute threads across all 20 CPU cores to maintain efficiency during tasks ranging from baseline productivity to local code debugging. * Microsoft Power and Thermal Framework (MPTF): Enabled natively on RTX Spark, MPTF standardizes thermal profiles across OEM designs, ensuring the systems maintain high sustained performance while preserving all-day battery life. * Graphics and AI APIs: The platform utilizes DirectX 12 advancements for neural rendering and optimized ray tracing, while Windows ML has been tuned to expose NVIDIA TensorRT capabilities natively to AI developers. Unified Memory Optimizations To fully leverage the 128GB unified memory pool, Windows has updated its memory allocation boundaries. The OS now implements a higher, intelligent limit on the total system memory accessible by the GPU. This adjustment allows users to load larger local large language models (LLMs) -- up to 120-billion parameters with a 1-million-token context window -- and render expansive 90GB 3D scenes. Additionally, enhanced page-size management within shared memory regions allows the OS to dynamically scale page sizes to accelerate heavy workloads while maintaining data transfer flexibility between the CPU and GPU. Prism Emulation Enhancements For legacy x86 compatibility, Microsoft's Prism emulator handles 32-bit and 64-bit applications on the Arm-based architecture. Building on previous updates that added AVX/AVX2 instruction set extensions, Prism has been structurally tuned for the specific microarchitecture of the RTX Spark superchip to close the performance gap for non-native development, creation, and gaming applications. Secure Architecture for On-Device AI Agents Addressing the security and privacy limitations of cloud-dependent AI, Microsoft and NVIDIA used Microsoft Build to showcase a local deployment platform for autonomous agents. The security architecture operates via native Windows primitives providing OS-enforced identity, containment, policy, and manageability. Layered atop this foundation is the NVIDIA OpenShell™ runtime, which grants users control over what local data an agent can access, intelligently routes queries based on compliance boundaries, and obfuscates personal details before communicating with external cloud models. Open-source agent applications like Hermes Agent and OpenClaw are integrating these security primitives directly into their Windows software, enabling them to execute multi-app workflows, write code, and conduct semantic file searches entirely on-device. Future phases of the collaboration will introduce these agent experiences directly into the Windows taskbar UI. Software and Gaming Ecosystem Growth The RTX Spark platform launches with an established ecosystem of native applications optimized for Arm-based Windows devices: * Creative Applications: Adobe has rearchitected its core suite -- including Photoshop, Premiere, and the Substance 3D tools (Painter and Stager) -- to run natively. The software utilizes a new video pipeline that couples unified memory with TensorRT to accelerate tools like Firefly-powered Generative Fill and Generative Extend by up to 2x. Other native creative tools include Blender 5.3 (featuring DLSS 4.5 Ray Reconstruction), DaVinci Resolve, Maxon Cinema4D, Topaz Photo, CapCut, and Affinity by Canva. * Technical Development: Engineering tools like MATLAB now officially support Windows on Arm. For AI developers, the platform introduces support for CUDA-accelerated PyTorch, llama.cpp, Hugging Face frameworks, Unsloth, Kohya, and node-based UIs like ComfyUI (featuring RTX Video with 4x Frame Generation). * Gaming Catalog: Native anti-cheat integrations from Epic's Easy Anti-Cheat and BattlEye, alongside Xbox PC app integration, allow x86 games to run smoothly under the enhanced Prism emulator. Game developers are bringing major titles natively to the platform, including Riot Games' League of Legends and VALORANT, KRAFTON's PUBG: Battlegrounds, as well as Alan Wake 2, Naraka: Bladepoint, Pragmata, and War Thunder. Hardware Form Factors and OEM Configurations Joining the Copilot+ PC category, RTX Spark systems combine an integrated NPU for baseline AI processing with the Blackwell GPU for advanced workloads. Microsoft Surface Laptop Ultra will be one of the first laptops with the chip. Initial laptop and compact desktop designs will debut this fall from major hardware manufacturers: Jensen Huang, founder and CEO of NVIDIA, said: The PC is being reinvented. For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask -- and the PC does the work. RTX Spark brings everything NVIDIA has built -- CUDA, RTX, our AI platform -- into a single superchip. Local agents. Frontier models. Creative workflows. RTX games. All on a laptop. This is the new PC. The personal AI computer.

First RTX Spark Laptops Wave Arriving This Fall, On-Stage Units Shown To Run AAA Games Like 007 First Light And Forza Horizon 6 On Battery Power

Several notebook manufacturers have teamed up with NVIDIA to bring forth various RTX Spark laptops to the market later this year. During the official announcement, the company stated that the chipset can be found in machines sporting Max-Q designs and a thinner form factor, with the increased memory count and the unified RAM bandwidth of 600GB/s able to seamlessly run AI models. However, perhaps the most impressive bit was that the RTX Spark has been demonstrated to run gaming titles like 007 First Light and Forza Horizon 6. Thanks to NVIDIA's Blackwell architecture and the help of DLSS and Frame Generation, it's possible to run AAA games, but there are a few details that need to be discussed In one of the presentations, NVIDIA mentioned that the RTX Spark can run AAA games at the 1440p resolution while maintaining 100FPS, which are impressive statistics. One of the best features of the chipset is its unified memory architecture, and with support of up to 128GB of LPDDR5X RAM, the RTX Spark isn't going to suffer from any VRAM problems when cranking up those visual settings, but what about the actual graphics performance? Looking at the computing units that NVIDIA founder and CEO Jensen Huang held during the keynote, both 007 First Light and Forza Horizon 6 have been able to run fluidly. However, this display of seamless performance was likely boosted by enabling DLSS and Multi-Frame Generation, which the RTX Blackwell GPU supports. Given that the RTX Spark won't suffer from VRAM limitations, supported titles can run 4x Multi-Frame Generation to boost the framerate, without experiencing stuttering or performance hitching. Then again, there's no mention of what graphics settings the two games were running at, which is extremely important context. During Apple's Cyberpunk 2077 demo, it was revealed that the M4 Max could run the game at 120FPS, but having tried out a multitude of GPUs, we were confident in our reporting that it wasn't possible to achieve this framerate unless there was upscaling and interpolation involved. Then again, Apple Silicon and RTX Spark are completely different architectures, so it'll be interesting to see the gaming results and other benchmarks when the first units arrive. As for the laptop partners, NVIDIA has said that Acer, ASUS, Lenovo, Dell, HP, Microsoft, and MSI will introduce their laptops later this year, but sadly, there's no mention of a starting price, nor do we have any idea regarding their base configuration. Fortunately, we'll keep readers updated in the future. Follow Wccftech on Google to get more of our news coverage in your feeds.

Nvidia Announces the RTX Spark, Hoping to Ignite a New Era of Agentic AI Computing. It Also Plays Games.

Nvidia finally announced the RTX Spark at Computex 2026, an SoC powering a 'new generation' of PCs purpose-built for Agentic AI, and basically every mainstream laptop manufacturer will be launching a laptop powered by it. Just like the DGX Spark, which made its debut at CES 2025, the RTX Spark is an SoC that's built for agentic AI. The chip will sport a 20-core ARM CPU, a Blackwell GPU with 6,144 CUDA cores and up to 128GB of VRAM that's shared between the two. Nvidia claims that this chip will boast up to 1 petaflop of AI compute power, and will be able to play games at 1440p with up to 100 fps. Nvidia is hoping that the RTX Spark, along with Windows, will bring about a new type of computing. Microsoft has been hinting at this type of thing in a while, but at Nvidia's press conference in Taipei, CEO Jensen Huang spelled it out himself, saying "For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask -- and the PC does the work." I'm not sure I'm ready for an era of shouting at my computer to try and make it do things, but it's not like I don't already do that anyways. It's also important to keep in mind that while the GPU in the DGX Spark looks impressive on paper, it will be stapled to an ARM processor. While it's likely that Nvidia will be able to work out the compatibility quirks with its nearly unlimited resources, ARM processors haven't traditionally been great for PC gaming. Even with the launch of Copilot+ laptops back in 2024, there were a ton of games that I couldn't even get to run on Qualcomm's chips, because they were built for an x86 platform. Only time will tell if Nvidia is able to work around that problem, but even if it does, I wouldn't exactly go out and get a RTX Spark-powered system as a gaming laptop. Instead, these systems are going to lean more into creative and AI workloads, and will likely have price tags to match. Neither Nvidia or any of the manufacturers that will be making laptops and mini PCs with the RTX Spark have announced prices, but its safe to assume that they'll be quite expensive - especially if they come with that full 128GB of memory. According to The Verge, Nvidia claims that there will be less expensive versions coming at some point with just 16GB of RAM, but it's not clear how long that wait will be. Either way, laptops powered by RTX Spark should be making their way to store shelves in Fall 2026, so if you do want to buy into Nvidia's vision of an Agentic future, you won't have to wait too long. Jackie Thomas is the Hardware and Buying Guides Editor at IGN and the PC components queen. You can follow her @Jackiecobra

Computex 2026: Microsoft Surface Laptop Ultra launched, check specs and features

The laptop is launching later in 2026 with no pricing announced yet Computex 2026: Microsoft has announced the Surface Laptop Ultra, its high-end laptop built around NVIDIA's newly announced RTX Spark chip. It is the first Surface laptop to feature an NVIDIA Blackwell RTX GPU with full CUDA support and Microsoft is positioning it squarely at developers, AI builders and creative professionals who have outgrown what conventional laptops can handle and it launches later this year. The RTX Spark chip at its heart combines a Blackwell RTX GPU with a 20-core Grace CPU, connected through NVIDIA's NVLink chip-to-chip interface and supports up to 128 GB of unified memory. That memory pool is dynamically shared between the CPU and GPU depending on workload, which means large AI models, heavy 3D renders and multi-model workflows can all run simultaneously without hitting memory walls. At the top end, the chip delivers 1 petaflop of AI compute and can run models with up to 120 billion parameters locally. For context, that puts locally-run AI at a scale most people have only been able to access via cloud services. You can read more about RTX Spark here. Microsoft Surface Laptop Ultra Specs and Features The Surface Laptop Ultra features a 15-inch mini-LED PixelSense Ultra touchscreen with a 3:2 aspect ratio and up to 2,000 nits of peak HDR brightness. Microsoft claims it is the brightest display it has ever shipped on a Surface. The display is designed with colour-critical work in mind and is calibrated for professionals making exposure and grading decisions. The chassis is under 18 mm thick and weighs under 2 kg, which is notable given the hardware inside. The touchpad is more than 30% bigger than the previous generation with haptic feedback. I/O Ports include: HDMI, USB-C, USB-A, a full-size SD card reader and a headphone jack, so there is no need for a dongle in most workflows. The laptop will be available in two colours: Platinum and a new Nightfall finish. Microsoft has also built in a user-replaceable SSD, which it frames as a longevity and IT management feature. The Surface Laptop Ultra is optimised for RTX Spark and runs on Windows 11, with Microsoft making specific OS-level changes to take advantage of the unified memory architecture, including raising the memory ceiling available to the GPU. Adobe has rearchitected Premiere Pro and Photoshop for RTX Spark with a new video pipeline in Premiere targeting real-time editing and GPU-accelerated colour correction, and a next-generation engine in Photoshop for GPU-accelerated compositing and live filters. Blender, DaVinci Resolve, Cinema 4D, CapCut, Topaz Photo and Affinity by Canva all run natively on the platform. For software still dependent on x86, Microsoft's Prism emulator now supports RTX Spark's GPU. On the gaming side, League of Legends, Valorant, PUBG, Alan Wake 2 and others have confirmed support. Since RTX Spark is a Windows on Arm platform, game compatibility will depend on a mix of native Arm builds, Prism emulation and anti-cheat support. The Surface Laptop Ultra will be available later in 2026 and pricing is yet to be announced.

NVIDIA RTX Spark brings CUDA, Blackwell and local AI agents to thin Windows laptops

Adobe, ASUS, Dell, HP, Lenovo, Microsoft Surface and MSI are among the early partners. For the last two years, the phrase "AI PC" has mostly meant one thing: a laptop with an NPU (Neural Processing Unit) capable of running some on-device AI tasks. Microsoft's CoPilot program has been quite successful at flooding the market with a bevy of AI-capable machines but it was always the CPU package doing most of the heavy lifting. NVIDIA, whose AI stack currently has the largest market share had to jump into the fray. And NVIDIA's new RTX Spark is an attempt to do just that. Unveiled at NVIDIA GTC Taipei, NVIDIA RTX Spark is being positioned as a new class of Windows PC chip built not merely for AI features, but for personal AI agents that can work locally, reason across apps, generate content, assist developers and still run games with the RTX stack. The company's framing is deliberately dramatic. NVIDIA says RTX Spark is designed for a world where the PC moves "from tool to teammate". Jensen Huang, founder and CEO of NVIDIA, put it even more directly, "The PC is being reinvented. For forty years, you launched apps. Click. Type. With RTX Spark and Microsoft Windows, you ask -- and the PC does the work. RTX Spark brings everything NVIDIA has built -- CUDA, RTX, our AI platform -- into a single superchip. Local agents. Frontier models. Creative workflows. RTX games. All on a laptop. This is the new PC. The personal AI computer." The underlying product is a 1-petaflop AI superchip for thin Windows laptops and compact desktops, combining a Blackwell RTX GPU, a 20-core Grace CPU, NVLink chip-to-chip connectivity, up to 128 GB of unified memory and the full NVIDIA software stack. The bigger question is whether this can do for AI PCs what discrete GeForce GPUs did for gaming laptops, i.e. create a performance category that users can understand, and software developers can reliably target. A different kind of AI PC Most current AI PCs are built around a CPU, integrated graphics, and an NPU that handles low-power AI workloads. RTX Spark takes a different route. NVIDIA is putting its GPU-first AI stack at the centre of the machine and using unified memory to make much larger local workloads possible. According to NVIDIA, RTX Spark features a Blackwell RTX GPU with 6,144 CUDA cores and fifth-generation Tensor Cores with FP4 precision. That GPU is connected through NVIDIA NVLink-C2C to a 20-core Grace CPU. MediaTek collaborated with NVIDIA on the custom CPU design, contributing to the Arm-based SoC's power efficiency, performance and connectivity. The chip is built to bring NVIDIA technologies such as CUDA, RTX, TensorRT, OptiX, DLSS, Reflex and G-SYNC into slim Windows laptops and small desktop PCs. The memory configuration is one of the most important parts of the story. RTX Spark supports up to 128 GB of unified memory, giving the GPU access to a much larger memory pool than typical laptop discrete GPUs. NVIDIA says this will allow users to run 120-billion-parameter large language models locally with up to 1 million tokens of context, render 90GB 3D scenes, edit 12K 4:2:2 video and generate 4K AI videos. NVIDIA said RTX Spark's top configuration uses 300 GB/s memory bandwidth and a 600 GB/s NVLink chip-to-chip interface. The company also said RTX Spark's graphics performance sits in the same broad class as an RTX 5070 laptop GPU, though this comparison will vary depending on workload because the architecture is different from a conventional x86 CPU plus discrete GPU design. The efficiency pitch is just as important as the peak-performance pitch. NVIDIA says RTX Spark laptops will be as slim as 14 mm and as light as three pounds, while still offering all-day battery life. In response to journalist questions, NVIDIA said light workloads can run at low single-digit wattages, while top laptop configurations can scale up to around 80 W under heavy workloads. It also said gaming battery life will vary heavily by frame-rate caps, settings, battery size and workload, and acknowledged that pulling maximum performance from any laptop battery can still drain it quickly. Local agents are the real platform play The most interesting part of RTX Spark is not that it is another high-performance chip. It is that NVIDIA is using it to push a new model of personal computing built around local agents. NVIDIA argued that AI agents have reached an inflection point, helped by the growth of open-source agent projects such as OpenClaw and Hermes Agent. The company's claim is that local models are now capable enough to power agents that can understand intent, use tools, work across applications and act proactively. NVIDIA frames the challenge more practically by stating that a broad adoption of agents has been limited by the difficulty of running them securely and privately on users' primary PCs. To address this, NVIDIA and Microsoft are collaborating on a Windows platform for on-device agents, built around new Windows security primitives and NVIDIA OpenShell. These Windows primitives are meant to provide identity, containment, policy and end-to-end security for agents. NVIDIA OpenShell adds another layer of user-defined policy, letting users decide what agents can and cannot do, which files and resources they can access, which queries must remain local, and when a cloud model can be used. NVIDIA says OpenShell can also disguise personal information in queries sent to cloud models. This matters because the agents NVIDIA is describing are not limited to a chat window. They may control apps, search local files, code plug-ins, generate images and videos, operate across workflows, and potentially use the screen, mouse and keyboard much like a user. The security and containment layer is therefore central to whether users and enterprises will trust these systems. Satya Nadella, chairman and CEO of Microsoft said, "Our goal is to deliver unmetered intelligence to every home and every desk with Windows. RTX Spark marks a real breakthrough towards that vision." NVIDIA also said Windows agent experiences powered by RTX Spark will eventually be accessible from the Windows taskbar interface. That suggests Microsoft and NVIDIA are thinking beyond third-party apps and towards agents that feel more native to Windows itself. What agents could actually do on a Spark laptop For creators, the company described an agent that can help an artist build complex generative AI workflows from a sketch, a mood board and natural-language instructions. Instead of forcing the user to master node-based AI tools, the agent could assemble the workflow, generate the image, produce additional camera angles and then call ComfyUI APIs to animate the output. The user could remain inside familiar tools such as Photoshop while the agent handles the more mechanical parts of the process. When it comes to developers, NVIDIA described agents that can monitor GitHub repositories and websites, identify open issues, propose fixes, test them using computer-use skills and prepare changes for approval. The company repeatedly stressed that the developer remains in command, but the agent can work through repetitive investigation and QA tasks, including overnight. And for gamers and streamers, the examples were simpler but more immediately understandable. A local agent could optimise monitor settings for esports, adjust game settings, configure peripherals, or manage a streaming setup by muting the microphone, switching an OBS scene and turning off lights when the streamer steps away. These are not grand, science-fiction tasks. They are the kind of small cross-app actions that are often tedious precisely because they involve several tools and settings panels. NVIDIA is also working on performance improvements for agent workloads. The company said it has collaborated with llama.cpp on multi-token prediction, a technique where a smaller draft model proposes multiple future tokens and a larger model verifies them in a single pass. NVIDIA claims this can deliver 2x performance in dense models and 1.6x performance in mixture-of-experts models while maintaining accuracy. The company has also worked with H Company to optimise the Holo computer-use model, claiming a 2x speed-up on NVIDIA GPUs and a 35 percent reduction in memory consumption. Georgi Gerganov, founder of llama.cpp said, "RTX Spark laptops change the game by multiplying the amount of context processing and putting it directly into a beautiful, portable chassis. Highly optimized models running locally through llama.cpp with RTX Spark's AI performance will unleash the next wave of personal, private agents." Adobe is reworking Premiere and Photoshop for Spark NVIDIA is not relying only on hypothetical agent workflows. The company is also pointing to major software vendors as proof that RTX Spark has a real application ecosystem behind it. The biggest creative announcement is Adobe. NVIDIA says Adobe is rearchitecting Premiere and Photoshop for RTX Spark to deliver up to 2x faster AI, editing, colouring and effects across creative workflows. Premiere will get a new video pipeline designed to use RTX Spark's unified memory, Blackwell GPU and TensorRT software for real-time editing, colour correction, GPU-accelerated AI and more efficient rendering of complex timelines. Photoshop is also getting a next-generation engine optimised for GPU-accelerated compositing, live filters, HDR and natural brushing. NVIDIA says the AI-native pipeline is being built to use TensorRT and the full power of RTX Spark. Adobe Substance 3D Painter and Stager will also run natively on RTX Spark for smoother 3D texturing and scene creation workflows. Shantanu Narayen, chair and CEO of Adobe, said: "The best creative work in the world happens in Adobe tools from Adobe Firefly to Photoshop and Premiere, and the expansion of our partnership with NVIDIA and Microsoft will make those experiences faster and more powerful than ever. Together, we are building AI-native creative experiences for RTX Spark that deliver the performance, intelligence and responsiveness people need to create at the pace of their ambition." NVIDIA also said Adobe will extend Premiere and Photoshop to work with Windows agents, so creators can create, edit and design with an AI teammate. Updates to Premiere, Photoshop and Substance are expected to start rolling out alongside RTX Spark availability. The broader partner list includes Adobe, Blackmagic Design, Blender, CapCut, ComfyUI and OTOY. On the gaming side, NVIDIA named KRAFTON, NetEase, Remedy Entertainment, Riot Games and Xbox among developers and providers embracing RTX Spark. Gaming support and the Windows on Arm problem NVIDIA is making it clear that RTX Spark is not just for AI developers and creators. It has the power of an RTX 5070 laptop GPU. The company says users will be able to play AAA games at 1440p and over 100 FPS with ray tracing, DLSS and Reflex. NVIDIA specifically mentioned titles such as Fortnite, Cyberpunk, Doom and Indiana Jones. The complication is that RTX Spark is a Windows on Arm platform. That means game support depends on a mix of native Arm versions, optimised ports, emulation through Microsoft's Prism layer, driver tuning and anti-cheat compatibility. NVIDIA said it is working with game developers and both first-party and third-party anti-cheat providers, but it did not name all anti-cheat partners but stated that they were working with all the big ones. We also did not get to see specific Prism overhead numbers or game benchmark comparisons against x86 systems. The company did say it has been working with Microsoft for years to improve Prism, including adding AVX2 support, which is important for some games and anti-cheat systems. NVIDIA also said there will be Game Ready drivers tuned for RTX Spark, and that some developers are working on native Windows on Arm versions while others are optimising for Prism. For gamers, this will be one of the biggest areas to watch. NVIDIA has a strong track record with PC gaming software, but Windows on Arm has historically struggled with compatibility. RTX Spark may have the GPU muscle, but the gaming experience will depend heavily on how many popular titles and anti-cheat systems behave well at launch. DLSS 4.5, Blender and AI-generated video RTX Spark will also debut or benefit from new RTX capabilities. NVIDIA announced DLSS 4.5 Ray Reconstruction with a second-generation transformer model, coming to Blender 5.3 and dozens of games. Ray Reconstruction, unveiled some time back, is designed to improve the quality of ray-traced scenes by replacing traditional denoisers with an AI model that better reconstructs pixels between sampled rays. NVIDIA said DLSS 4.5 Ray Reconstruction uses a more efficient denoiser with 35 percent more compute capability and 20 percent more parameters, while maintaining similar performance on RTX GPUs, including Turing and Ampere. It is expected to be available in August for all RTX GPUs, and on RTX Spark when the platform launches. Blender support is a notable creator-side addition. NVIDIA says DLSS Ray Reconstruction in Blender 5.3 will improve interactive path tracing in the viewport, helping artists assess scenes in real time rather than waiting for frames to resolve. ComfyUI will also get RTX Video with 4x Frame Generation, aimed at making AI-generated video smoother and longer than the short 15 to 20fps clips common today. Yannik Marek, co-founder and creator of ComfyUI, said: "The combination of RTX Spark's processing capabilities and large unified memory will make it one of the best-performing laptops to run diffusion models. ComfyUI users can now run highly complex, multimodal workflows and generate ultra-high-resolution images and videos with unprecedented speed on a portable device." What still needs proving RTX Spark is one of NVIDIA's most ambitious PC announcements in years, but several details remain open. NVIDIA did not disclose pricing, full SKU breakdowns, detailed CPU benchmarks, exact gaming performance, emulator overhead, comparative figures against Apple's M-series, AMD Ryzen AI, Intel Core Ultra or Qualcomm Snapdragon X platforms, or precise battery-life figures under heavy AI and gaming workloads. The company repeatedly said more performance details would be shared closer to availability. There are also product segmentation limits. ECC memory support is reserved for Pro products. NVIDIA said the current announcement is focused on consumer systems, not commercial notebooks with vPro-like management features. It also said RTX Spark is a Windows platform, while DGX Spark remains aimed at Linux AI developers. The two products use related silicon, but NVIDIA described them as separate platforms with different software targets. That leaves RTX Spark in an interesting place. It is not a conventional gaming laptop chip. It is not merely an NPU-led AI PC processor. It is not DGX Spark for Linux developers either. It is NVIDIA's attempt to create a Windows machine where local agents, CUDA workflows, RTX gaming, creator apps and unified memory all converge. The timing also makes sense. The first wave of AI PCs has struggled to communicate why ordinary users should care about NPUs. RTX Spark offers a clearer story for power users who can now run bigger models locally, handle heavier creative workloads, use agents with more context, and still play games. The risk is that this depends on many moving parts working at once: Windows on Arm, Prism, anti-cheat support, native apps, Adobe's new engines, OpenShell, local agent maturity and OEM pricing. If NVIDIA and Microsoft can make that stack feel seamless, RTX Spark could become a meaningful shift in the Windows laptop market. If not, it may remain a technically impressive platform waiting for the software ecosystem to catch up. Either way, it changes the AI PC conversation from "how many TOPS does the NPU have?" to a more relevant question: what can a personal computer actually do when it has enough local AI performance, enough memory and an agent that can safely act on the user's behalf? RTX Spark laptops and desktops arrive this fall RTX Spark laptops and compact desktops will be available this fall i.e. from ASUS, Dell, HP, Lenovo, Microsoft Surface and MSI, with Acer and GIGABYTE models to follow. NVIDIA named several early laptop designs, including ASUS ProArt P14 and P16, Dell XPS, HP OmniBook X 14, Lenovo Yoga Pro 9N, Microsoft Surface Ultra and MSI Prestige N16 Flip AI+.

New NVIDIA SoCs powered Windows laptops coming: Why it matters

A serious NVIDIA-ARM computing challenge to Intel, AMD, Qualcomm All of us who thought NVIDIA had abandoned the consumer tech segment to focus more efforts on making a killing in the AI and datacentre market, Computex 2026 promises to bring some much-needed good news. According to leaks and reports, it looks like NVIDIA is almost ready to make a big announcement that will put a smile on Windows laptops users. And no, it doesn't look like another glorified GPU inside a mobile computing form factor. In all likelihood, it looks like an NVIDIA-designed ARM SoC that will power the whole Windows 11 PC in the near future. Leaked reports suggest the expected launch of NVIDIA N1 / N1X, an ARM-based Windows laptop platform developed with MediaTek's help, which will have NVIDIA's Blackwell GPU architecture integrated inside the SoC. Although unconfirmed right now, leaked information suggests the soon-to-launch NVIDIA N1X chip uses Cortex-X925 and Cortex-A725 cores, Blackwell graphics, and power envelopes from 45-80W for N1X and 18-45W for N1. Just based on this, it looks like the new NVIDIA chips-based Windows 11 laptops aren't meant to be for thin-n-light laptops like the Snapdragon X-series. They seem to be geared towards Apple's M-series workloads. Also read: AI is everywhere but handheld gaming could steal the show at Computex 2026 According to The Verge, NVIDIA's N1 and N1X laptop chips-based laptop announcements are expected from the likes of Lenovo and Dell very soon. So far, on Windows 11 PCs, NVIDIA has never been the main component, always the side-dish served next to Intel and AMD-based CPU configurations. That is what makes this upcoming N1 and N1X announcement so different. This isn't NVIDIA trying to smuggle itself into the PC through a graphics engine. Now, NVIDIA is attempting to become the PC's singular, unified platform itself - kinda like how Qualcomm and Apple have done with their respective laptop sojourns so far. For NVIDIA, this isn't their first Windows PC rodeo. They've been here before, more than a decade ago, with failed adventures. With Project Denver, NVIDIA's 2011 CES declaration to make high-powered ARM CPU cores from everything, Windows PCs to supercomputers. Who can forget Surface RT powered by the NVIDIA Tegra 3 chip? It was ARM before it was ready, Windows before it was ready, overall a cautionary tale. But today, in 2026, the situation is much different. Windows on ARM is no longer a punchline for failure. Microsoft has shown it's possible, as Qualcomm pioneered the category into legitimacy. Not to forget Apple here, which has proved that consumers don't care about their laptop's instruction sets if the machine is fast, quiet and lasts long enough to make the charger feel optional. Also, NVIDIA isn't merely asking whether Windows laptops need better graphics. Looks like it's attempting to see if the modern PC should be organised around AI compute, unified memory, ARM efficiency, RTX graphics and CUDA familiarity from day one. Of course, plenty of questions still remain. But this feels like NVIDIA's most decisive PC-chip attempt ever. N1X, if it lands, isn't trying to fit into the mould of the Windows PC that it was, because surely it will be aimed squarely to dictate where the PC is heading next. Intel, AMD and Qualcomm better watch out.