Nvidia Unveils RTX Spark Superchip for PCs and Laptops with AI Agents

Nvidia officially unveiled the RTX Spark platform, designed for Windows-based laptops and desktops. The key feature of the new product is the integration of AI agents capable of performing tasks in the background, analyzing context, and interacting with users through natural language. To achieve this, Nvidia leveraged Arm architecture, combining it with a Blackwell GPU and 128GB of LPDDR5X memory.

The RTX Spark superchip is equipped with a 20-core CPU, 6,144 CUDA cores in the GPU, and delivers memory bandwidth of up to 300GB/s. According to Nvidia, this configuration will enable the execution of models with 120 billion parameters and process contexts up to one million tokens long. The platform supports NVLink C2C technology for rapid data exchange between CPU and GPU, which is critical for AI agent performance.

The first devices powered by RTX Spark are expected to launch in autumn 2026. Nvidia’s partners—Dell, HP, Lenovo, Microsoft, Asus, and MSI—will introduce laptops featuring OLED displays, G-Sync support, and premium design. Nvidia also announced compact desktop systems, including the DGX Spark model. The platform is positioned as a versatile solution for gamers, creative professionals, and AI developers.

In collaboration with Adobe, Nvidia is optimizing Photoshop and Premiere for RTX Spark, transitioning them to full GPU acceleration. This will accelerate generative workflows and ultra-high-resolution video processing. Meanwhile, Microsoft is integrating the OpenShell framework and new security mechanisms into Windows to control AI agents' access to user data.