Nvidia Launches PAIR, a Free Tool That Distributes Local AI Inference Across Home GPUs

Nvidia announced Personal AI Router (PAIR) on Sept 3 at IFA 2026, a free open-source tool that auto-discovers compatible PCs on a home network and routes inference requests to whichever GPU has capacity — enabling parallel agentic workflows entirely offline. Beta supports Windows, macOS, and Linux with GeForce RTX 20-series and newer, RTX PRO workstation GPUs, DGX Spark, and Apple M4 silicon, integrating with Ollama and LM Studio at launch. Nvidia paired the reveal with an October ship window for the RTX Spark N1X in two configurations (18 or 20 CPU cores, 5,120 or 6,144 Blackwell CUDA cores, up to 128GB unified memory).