NVIDIA RTX Spark Officially Revealed as New Windows on ARM Chip for AI Laptops
NVIDIA has officially introduced RTX Spark, the consumer-focused chip previously rumored under the N1 and N1X names. Built for the next generation of Windows on ARM laptops, RTX Spark marks a major step for NVIDIA as it moves deeper into the PC processor market with a platform designed for AI, gaming, creative workloads, and high-performance computing.
The chip is the result of NVIDIA’s partnership with MediaTek, combining a powerful Grace-based CPU with RTX Blackwell graphics technology. NVIDIA is positioning RTX Spark as a major shift in personal computing, where laptops are no longer just devices for launching apps but intelligent systems capable of running advanced AI tasks locally.
During the GTC Taipei keynote, NVIDIA founder and CEO Jensen Huang described RTX Spark as a new kind of PC platform that brings together the company’s biggest technologies, including CUDA, RTX graphics, and its AI software stack. The idea is to create a laptop that can handle local AI agents, large models, creative workflows, and RTX gaming without relying entirely on the cloud.
One of the biggest highlights of RTX Spark is its CPU design. The chip can feature up to a 20-core Grace CPU, developed with support from MediaTek. NVIDIA claims this setup will deliver strong single-core and multi-core performance for Windows on ARM laptops. However, early benchmark results suggest there is still room for improvement, especially when compared with established chips such as Apple’s M3 Max. Final performance will depend heavily on retail hardware, cooling designs, software optimization, and driver maturity.
On the graphics side, RTX Spark includes a Blackwell-based RTX GPU with 6,144 CUDA cores. NVIDIA says the chip can deliver up to 1 petaFLOP of AI performance, making it a serious contender for users who need portable machines capable of handling demanding AI and graphics workloads. Some early expectations place the GPU performance near a laptop-class RTX 5070, though real-world testing will be needed before making firm comparisons.
AI is clearly at the center of RTX Spark’s identity. NVIDIA says the platform is built to run massive AI workloads, including models with up to 120 billion parameters and a 1 million token context window. That could make future RTX Spark laptops useful for developers, researchers, creators, and professionals who want powerful on-device AI without depending on remote servers.
For gamers and creators, the use of RTX Blackwell technology should bring support for modern graphics features, AI-enhanced rendering, and accelerated creative applications. With CUDA support included, RTX Spark could also appeal to developers and technical users who already rely on NVIDIA’s ecosystem for machine learning, simulation, video editing, 3D rendering, and other GPU-accelerated workloads.
The arrival of RTX Spark also adds new momentum to the Windows on ARM market. While ARM-based Windows laptops have improved in recent years, NVIDIA’s entry could make the platform far more attractive, especially if software compatibility, battery life, thermals, and performance all come together in shipping devices.
A large wave of RTX Spark-powered laptops is expected later this year. If NVIDIA and its partners can deliver strong performance, efficient power usage, and broad app support, RTX Spark could become one of the most important laptop chip launches in the Windows ecosystem.
For now, RTX Spark looks like NVIDIA’s bold attempt to redefine the AI PC. With a 20-core Grace CPU, Blackwell RTX graphics, thousands of CUDA cores, and serious AI processing capability, it has the potential to bring workstation-class features into thin and portable laptops. The real test will come when commercial devices arrive and users can see how well NVIDIA’s new Windows on ARM platform performs in everyday use, gaming, creative work, and local AI tasks.
