ASUS Debuts ProArt RTX Spark Series for On-Device AI Workflows

The new hardware centers on the RTX Spark architecture, which integrates a Blackwell-class GPU featuring 6,144 CUDA cores with a 20-core NVIDIA Grace CPU. This configuration supports up to 128GB of unified memory, allowing developers to execute 120B-parameter large language models and render 90GB 3D scenes without the latency or token constraints of cloud-based services. ASUS claims the move toward local processing ensures complete data privacy for professional creative workflows.

Design-wise, the P16 and P14 represent the slimmest entries in the ProArt series, weighing 1.77kg and 1.48kg respectively, while housing high-capacity 99Wh batteries. Each device features an ASUS Lumina Pro OLED display, calibrated for Delta E < 1 color accuracy with 1,600 nits of peak brightness. For stationary setups, the GR1X mini PC condenses this power into a 150 x 150 x 51mm chassis optimized for continuous AI processing. The ecosystem is further bolstered by proprietary software like MuseTree and StoryCube, which integrate generative models such as FLUX and WAN to support agentic tasks like the Zenni Claw workflow.