Nvidia and Microsoft have unveiled a new class of Windows PCs powered by Nvidia chips, marking one of the clearest signs yet that the personal computer market is being pulled deeper into the AI hardware race.
The new machines are built around Nvidia RTX Spark, a new superchip designed to bring more AI processing directly onto laptops and desktop computers. Nvidia announced the platform ahead of the Computex trade show in Taiwan, with systems expected from Microsoft Surface and major PC manufacturers including Dell, HP, Lenovo, ASUS, MSI and others.
The launch moves Nvidia beyond its traditional strength in graphics processing units (GPUs) and further into the market for central processing units (CPUs), the main chips that run computers. For Microsoft, it also adds another route into AI-focused Windows PCs as the company continues trying to make local AI a more central part of the Windows experience.
Nvidia and Microsoft Push AI Onto the PC
Nvidia said RTX Spark is designed for Windows PCs built around personal AI agents. These are AI systems that can carry out tasks on a user’s behalf, such as handling creative workflows, assisting developers, managing files or working across applications.
The important shift is where that work happens.
Most AI tools today still depend heavily on cloud computing, where data is sent to remote data centres for processing. RTX Spark is designed to run more of that AI work locally on the device itself. That could make AI tasks faster, reduce cloud dependence and give users more control over sensitive data.
Nvidia CEO Jensen Huang said the PC is being “reinvented” for the AI era, with users moving from launching apps manually to asking the computer to complete work for them.
Microsoft is working with Nvidia to support that shift inside Windows. Nvidia said the companies are collaborating on a native Windows experience for personal agents, including new security controls and Nvidia OpenShell, which is designed to help agents run safely on primary devices.
For enterprise users, that security point matters. AI agents are only useful if they can interact with real files, applications and workflows. But the moment they do that, they also raise questions about access, containment and control. That’s why local AI on PCs isn’t only a performance story. It’s also a governance story.
What RTX Spark Brings to Windows PCs
RTX Spark combines several parts of Nvidia’s hardware and software stack into one chip platform. According to Nvidia, the superchip includes a Blackwell RTX GPU, a 20-core Grace CPU, fifth-generation Tensor Cores, and up to 128GB of unified memory.
In simple terms, unified memory lets the CPU and GPU access the same pool of memory instead of splitting resources between separate systems. That can help with large workloads, including AI models, video editing and 3D rendering.
Nvidia said RTX Spark can deliver one petaflop of AI performance and support local 120-billion-parameter large language models. A large language model (LLM) is the type of AI model used to understand and generate text, code and other forms of language-based output.
The company is positioning the chip for creators, developers, gamers and AI users who need more local processing power without moving entirely into workstation or cloud infrastructure.
MediaTek also played a role in the chip’s development. Nvidia said the Taiwanese chip designer collaborated on the custom CPU design, contributing to the platform’s power efficiency, performance and connectivity.
A New Challenge to Intel, AMD and Qualcomm
The launch puts Nvidia into direct competition with Intel, AMD and Qualcomm in the Windows PC market.
Intel and AMD have long dominated the market for Windows laptop and desktop processors. Qualcomm has been pushing Windows on Arm devices, using the same Arm-based architecture that powers many smartphones and Apple’s Mac chips.
Nvidia’s RTX Spark is also Arm-based. That means some older Windows applications built for x86 chips may need to run through emulation, where software translates instructions so the app can work on Arm hardware. Microsoft has been improving this through its Prism emulator, but app compatibility will still be one of the practical details buyers watch closely.
Reuters reported that Nvidia-powered computers are expected from Microsoft’s Surface brand and other computer makers including Dell. Nvidia has now said RTX Spark-powered slim Windows laptops and compact desktop PCs will be available this fall from ASUS, Dell, HP, Lenovo, Microsoft Surface and MSI, with Acer and GIGABYTE models to follow.
HP has also confirmed plans for RTX Spark-powered OmniBook laptops later this year, describing the platform as a way to support local AI workflows for creators, developers and gamers.
Why This Matters for the AI PC Market
The AI PC market has been building for some time, but adoption hasn’t yet produced the kind of mass upgrade cycle Microsoft and PC makers would like to see.
Nvidia entering the market changes the competitive picture. The company already has a strong position in AI infrastructure and graphics hardware. By moving into full PC chips, it’s trying to carry that AI advantage from data centres and workstations into everyday Windows machines.
That doesn’t mean RTX Spark PCs will immediately reshape the market. Pricing, battery life, app compatibility and real-world AI use cases will decide how much demand there is beyond early adopters.
But the direction is clear. Nvidia and Microsoft are betting that the next phase of Windows PCs won’t be defined only by faster processors or better displays. It’ll be defined by whether the PC can run useful AI locally, safely and efficiently. The first RTX Spark-powered Windows PCs are expected later this year.
Comments ( 0 )