Nvidia Announces RTX Spark Platform With 20 Cores for Local AI Agents

Nvidia announced the RTX Spark platform at Computex 2026, aiming to reinvent personal computers for the agentic AI era. The company partnered with Microsoft, Dell, HP, MSI, Lenovo, and Asus to deliver this hardware.

The Microsoft Surface Laptop Ultra serves as the initial testbed for this architecture. It contains the full RTX Spark Superchip alongside 128GB of LPDDR5X-9400 memory. This configuration supports secure, contained execution of local AI agents.

Platform Performance Context

Testing by Tom's Hardware revealed that the MediaTek Arm CPU complex is the platform's weakest link. In Geekbench evaluations, both single-core and multi-core CPU scores fell significantly behind Apple's M5 Max system. The M5 Max comparison unit featured 18 CPU cores and a 40-core GPU.

Gaming remains a secondary focus for this release. While the system handles regular workloads without feeling slow, it is not positioned as a high-performance gaming machine. Detailed reviews of other RTX Spark laptops from partners are expected soon.

Affected Organizations and Suppliers

Microsoft, Dell, HP, MSI, Lenovo, and Asus are partners in this initiative. They hope to pave the way for PCs that act as active, autonomous companions for work and creation. IT departments should monitor how Windows on Arm handles local AI workloads compared to traditional systems.

Defender's checklist

• Evaluate current software compatibility with Windows on Arm environments.
• Review local AI agent security policies for contained execution models.
• Monitor vendor updates for performance data on MediaTek Arm components.
• Compare total cost of ownership against existing Apple Silicon deployments.

Operational impact

Teams usually need to verify that legacy applications run correctly on new Arm-based architectures before deploying them widely. Check whether your current security tools support the specific Windows on Arm version used by this platform. Ensure that local AI agents operate within defined containment boundaries to prevent data leakage. Inform procurement and IT leadership about the shift from general-purpose computing to specialized AI companion devices. This change may require updates to existing hardware refresh cycles and software licensing agreements.

Background: AI agents

An AI agent is a software system that perceives its environment, makes decisions, and takes actions to achieve specific goals without constant human input. Unlike passive tools, agents execute tasks, browse data, and interact with other systems autonomously.

Read the full guide: AI Agents Defined: Autonomy, Risk, and Real-World Use

What to do and how to stay safe: Nvidia

• Audit existing software portfolios for compatibility with Windows on Arm.
• Define clear security boundaries for local AI agent execution.
• Monitor vendor performance benchmarks before large-scale procurement.
• Consult with IT vendors about migration paths from x86 to Arm.

Step-by-step guide: Implementing AI Usage Policies at Work: A Step-by-Step Framework

General security guidance from the InfoSec Ledger newsroom. It is not confirmed advice from the organisations named in this story.

Frequently asked questions

What are the CPU specifications of the RTX Spark Superchip?

The chip has 20 CPU cores and 6144 CUDA cores, paired with 128GB of LPDDR5X-9400 RAM.

How does the RTX Spark CPU performance compare to Apple's M5 Max?

It greatly trails the M5 Max in both single-core and multi-core Geekbench performance tests.

Which companies are partnering with Nvidia on this platform?

Partners include Microsoft, Dell, HP, MSI, Lenovo, and Asus.