--- title: Nvidia RTX Spark: what we know about the new ARM chip... description: Nvidia unveiled the RTX Spark (N1/N1X) at Computex 2026, an ARM-based chip combining a 20-core Grace CPU with a Blackwell GPU featuring 6,144 CUDA cores, 128GB... canonical: https://daily.dev/posts/nvidia-rtx-spark-what-we-know-about-the-new-arm-chip-coming-to-windows-laptops-and-desktops-xxguxsmqv twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Nvidia RTX Spark: what we know about the new ARM chip coming to Windows laptops and desktops | daily.dev og:description: Nvidia unveiled the RTX Spark (N1/N1X) at Computex 2026, an ARM-based chip combining a 20-core Grace CPU with a Blackwell GPU featuring 6,144 CUDA cores, 128GB... og:url: https://daily.dev/posts/nvidia-rtx-spark-what-we-know-about-the-new-arm-chip-coming-to-windows-laptops-and-desktops-xxguxsmqv og:image: https://api.daily.dev/og/posts/xxguXsmqV.png og:image:alt: Nvidia RTX Spark: what we know about the new ARM chip coming to Windows laptops and desktops og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Nvidia RTX Spark: what we know about the new ARM chip coming to Windows laptops and desktops **[Collections](https://daily.dev/sources/collections)** · 4 min read · 1 upvotes · 0 comments ## Summary Nvidia unveiled the RTX Spark (N1/N1X) at Computex 2026, an ARM-based chip combining a 20-core Grace CPU with a Blackwell GPU featuring 6,144 CUDA cores, 128GB unified LPDDR5X memory, and 1 petaflop of AI compute within a 45–80W envelope. It marks Nvidia's first integrated CPU+GPU chip for Windows PCs and brings the full CUDA stack to ARM Windows for the first time. Announced devices include Microsoft Surface Laptop Ultra, Surface RTX Spark Dev Box, Asus ProArt laptops and mini PC, MSI Prestige N16 Flip, and Nvidia DGX Windows workstation. Software support includes x86 Prism emulation, OpenShell for local AI agents, Microsoft eXecution Containers for sandboxing, and local LLM support for Llama, Mistral, Gemma, and Nvidia's Nemotron models. The chip positions Nvidia against Qualcomm, Intel, AMD, and Apple Silicon in the PC market, with gaming support also confirmed. Key unknowns remain: sustained thermal performance, battery life, and real-world x86 emulation compatibility. ## Content ## What Nvidia announced At Computex 2026, Nvidia unveiled the RTX Spark (also referred to as N1/N1X), an ARM-based superchip that pairs a 20-core Grace CPU — co-developed with MediaTek — with a Blackwell-generation GPU carrying 6,144 CUDA cores. The two are connected via NVLink-C2C, sharing up to 128GB of LPDDR5X unified memory at 600GB/s bandwidth. Total AI compute lands at 1 petaflop, within a 45–80W power envelope. This is the first time Nvidia has shipped a full CPU+GPU integrated chip for Windows PCs, and it brings the complete CUDA software stack with it — something no Windows on ARM chip has offered before. ## The hardware lineup Several devices were announced alongside the chip: - **Microsoft Surface Laptop Ultra** — 15-inch mini-LED PixelSense Ultra display (2,000-nit peak), up to 128GB unified memory, 4.5lbs, fall release - **Microsoft Surface RTX Spark Dev Box** — developer workstation pre-configured with WSL2 GPU passthrough, CUDA support, VS Code, and GitHub Copilot; can run 120B parameter models locally - **Asus ProArt P16** — 16-inch 4K Tandem OLED, 12.9mm thin, 1.77kg, 128GB RAM, 100% DCI-P3 - **Asus ProArt P14** — 14-inch 3K HDR 120Hz display - **Asus ProArt Mini PC** — 150×150×51mm chassis, 128GB unified memory, 10GbE networking - **MSI Prestige N16 Flip AI+** — thin convertible with confirmed gaming support - **Nvidia DGX Windows workstation** — desktop-class RTX Spark system - Devices from Dell, HP, and Lenovo are also expected this fall Some of the laptops are reportedly as thin as 14mm and as light as 1.4kg. Pricing hasn't been announced for most of them. ## The software side Microsoft introduced two OS-level features to support the chip: Workload Profile Scheduling (WPS) for performance and efficiency tuning, and the MPTF thermal framework. RTX Spark also includes x86 Prism emulation for legacy app compatibility — which addresses the obvious concern about ARM Windows and older software, though real-world compatibility at scale remains to be seen. For AI workloads, Nvidia and Microsoft co-developed two complementary tools: - **OpenShell** — a runtime for deploying autonomous AI agents locally, with privacy controls - **Microsoft eXecution Containers (MXC)** — secure sandboxing for agent workloads The chip is designed to run local LLMs including Nvidia's Nemotron models, as well as Llama, Mistral, and Gemma. The pitch is the usual local AI argument: better privacy, no cloud costs, faster responses, offline capability. Other software updates announced alongside RTX Spark include: - NVIDIA NemoClaw support across all RTX client systems - Hermes Agent native Windows desktop app - H Company's Holo 3.1 models with 35% lower memory footprint - 2x inference speedup on Qwen models in llama.cpp via Multi-Token Prediction - 2.6x gains in vLLM - Multi-GPU tensor parallelism in llama.cpp and ComfyUI ## Why this is a bigger deal than it looks Nvidia CEO Jensen Huang has framed RTX Spark as the company's entry into the $200B CPU market. That's not subtle. Nvidia has dominated AI compute in data centers; now it's making a direct play for the PC, putting it in competition with Qualcomm, Intel, AMD, and — more pointedly — Apple Silicon. The Apple comparison is hard to avoid. Grace + Blackwell in a laptop is essentially Nvidia's answer to six years of M-series dominance. Whether it actually competes depends on things we don't know yet: sustained performance under thermal constraints, battery life in real use, and how well x86 emulation holds up for the apps people actually run. The gaming angle is also worth noting. The MSI Prestige N16 Flip confirms gaming support, and there are unverified claims that RTX Spark performance could approach an RTX 5070. Native anti-cheat and DRM support for Windows on ARM is included. If that holds up, it changes the calculus for ARM Windows significantly — gaming has historically been one of the platform's weakest points. The 2013 Surface RT comparison is fair to raise. That experiment failed partly because the hardware was underpowered and the software ecosystem wasn't there. RTX Spark has far more compute, a real GPU, the full CUDA stack, and x86 emulation. The ecosystem problem is less severe than it was. But "less severe" isn't the same as "solved," and fall 2026 devices will be the real test. ## Similar posts on daily.dev - [Nvidia's RTX Spark is a chip unlike any other, and it could change Windows laptops forever](https://daily.dev/posts/nvidia-s-rtx-spark-is-a-chip-unlike-any-other-and-it-could-change-windows-laptops-forever-lgrsetnyj) · XDA Developers · 0 upvotes · 0 comments - [Microsoft built its most powerful Surface ever on Nvidia's new chip, and we saw every RTX Spark competitor coming this fall](https://daily.dev/posts/microsoft-built-its-most-powerful-surface-ever-on-nvidia-s-new-chip-and-we-saw-every-rtx-spark-comp-51dx7bfrq) · XDA Developers · 0 upvotes · 0 comments --- Tags: [#windows](https://daily.dev/tags/windows), [#nvidia](https://daily.dev/tags/nvidia), [#arm](https://daily.dev/tags/arm), [#ai-inference](https://daily.dev/tags/ai-inference) [View this post on daily.dev](https://daily.dev/posts/nvidia-rtx-spark-what-we-know-about-the-new-arm-chip-coming-to-windows-laptops-and-desktops-xxguxsmqv) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Nvidia RTX Spark: what we know about the new ARM chip coming to Windows laptops and desktops","url":"https://daily.dev/posts/nvidia-rtx-spark-what-we-know-about-the-new-arm-chip-coming-to-windows-laptops-and-desktops-xxguxsmqv","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/nvidia-rtx-spark-what-we-know-about-the-new-arm-chip-coming-to-windows-laptops-and-desktops-xxguxsmqv"},"datePublished":"2026-06-01T16:26:40.639Z","dateModified":"2026-06-02T23:40:19.216Z","description":"Nvidia unveiled the RTX Spark (N1/N1X) at Computex 2026, an ARM-based chip combining a 20-core Grace CPU with a Blackwell GPU featuring 6,144 CUDA cores, 128GB...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/34aae0c32484f0c7aee7b9f07693b88e?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/34aae0c32484f0c7aee7b9f07693b88e?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/nvidia-rtx-spark-what-we-know-about-the-new-arm-chip-coming-to-windows-laptops-and-desktops-xxguxsmqv","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"windows,nvidia,arm,ai-inference","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Nvidia RTX Spark: what we know about the new ARM chip coming to Windows laptops and desktops"}]} ```