Nvidia launches RTX Spark chip for AI PCs

Nvidia unveiled the RTX Spark PC chip at Computex Taipei, developed with MediaTek and after three years of collaboration with Microsoft. The chip is designed to run autonomous AI agents locally in laptops and desktops from major PC brands. Nvidia also highlighted its Vera CPU and early adopters including OpenAI, Anthropic and SpaceX. The move expands Nvidia from AI training hardware into local inference and agentic AI PCs.

• Chip debuts this fall in laptops and desktops from Dell HP Lenovo and others

• Nvidia pairs consumer-grade RTX Spark with server-class Vera CPU in dual-tier AI chip strategy

The fact

On June 1, 2026, Nvidia unveiled the RTX Spark PC chip at Computex Taipei. Developed with MediaTek and after three years of collaboration with Microsoft, it integrates AI capabilities locally into laptops and desktops from Dell, HP, Lenovo, ASUS, Microsoft Surface, and MSI, with Acer and GIGABYTE to follow. The chip is designed to run autonomous AI agents rather than relying solely on cloud computing. Nvidia also introduced its Vera CPU targeting a $200bn market, with early adopters including OpenAI, Anthropic, and SpaceX.

The Assessment

RTX Spark marks Nvidia's return to the PC processor arena, combining a custom Arm-based CPU with Blackwell GPU in a single SoC. The strategic target is not just gaming or AI inference but the Windows on Arm ecosystem itself — an area Intel and AMD have long dominated with x86. By providing a reference design and partnering with Microsoft and MediaTek, Nvidia is attempting to replicate its datacentre playbook: own the full stack from silicon to software.

For BTW readers, the move could accelerate the architecture shift away from x86 dominance in client computing, with implications for PC OEMs, chip supply chains, and enterprise endpoint standardisation.

What to Watch

Watch OEM pricing and fall launch timelines, Windows on Arm software compatibility, and whether AMD or Intel accelerate their own Arm-based PC chip roadmaps.