Nvidia CEO Jensen Huang has confirmed that RTX Spark is not a one-off chip, with the company already planning future versions called N2X and N3X.
The confirmation comes days after Nvidia introduced RTX Spark at Computex 2026, positioning the new chip as a platform for laptops and compact desktops built around local AI agents, creator workflows and gaming.
RTX Spark is designed for users who want to run advanced AI workloads directly on personal computers instead of relying fully on the cloud.
The chip supports up to 128GB of unified memory, allowing the CPU and GPU to share the same memory pool. This makes it more suitable for large AI models and agent-based workflows than traditional laptops with smaller dedicated graphics memory.
The platform combines Nvidia’s Arm-based Grace CPU with a Blackwell RTX GPU. The higher-end configuration includes up to 20 CPU cores, 6,144 CUDA cores, and up to 1 petaflop of AI compute performance.
Huang said Nvidia plans to expand the RTX Spark family beyond the first chip, which was internally known as N1X.
The next versions, N2X and N3X, are already planned, signaling that Nvidia sees RTX Spark as a long-term PC platform rather than a single product launch.
The company is pitching these systems as a new kind of personal computer built for AI agents that can keep working even when the user is away.
RTX Spark laptops and compact desktops are expected to launch this fall from Asus, Dell, HP, Lenovo, Microsoft Surface, and MSI.
Acer and Gigabyte models are expected to follow later.
Microsoft has already shown the Surface Laptop Ultra, which uses RTX Spark and is described as the most powerful Surface Laptop it has built. The device is aimed at creators, developers, and power users who need local AI performance in a portable form factor.
Huang described the future PC as a device users can interact with more naturally and continuously, rather than a machine that simply waits for input.
He compared the idea to sci-fi-style AI companions, suggesting that future computers could become more active assistants that users speak with and assign tasks to throughout the day.
That vision is central to RTX Spark. Nvidia wants these PCs to run local AI agents, handle complex workflows, and support personal assistant-like computing without sending every task to remote servers.
RTX Spark places Nvidia more directly into the Windows PC processor market, where it will compete with Intel, AMD, and Qualcomm.
The first generation will need to prove itself on performance, battery life, software compatibility, and real-world AI use cases. However, Nvidia’s decision to plan N2X and N3X already suggests the company is serious about building a longer-term role in personal computing.
If RTX Spark devices perform well when they arrive later this year, Nvidia could become a much bigger force in premium Windows laptops and small desktop PCs.