Nvidia has announced the Personal AI Router (PAIR), a free tool that connects home computers on a network to handle local AI inference tasks. Despite its name, PAIR is not a hardware router but open-source software developed by Nvidia. It discovers compatible PCs on a network, links them together, and prepares them to run agentic workflows.
PAIR is designed to work with local AI tools such as Ollama and LM Studio. Compatible devices are mostly Nvidia GeForce GPUs, including RTX 20-series cards and newer, RTX Pro GPUs, and DGX Spark systems. Apple’s M4 chips or newer are also supported.
Why it matters
The tool lets users pool idle computing resources across multiple machines for AI workloads, running inference locally rather than relying solely on cloud services.
Who should care
Users who run local AI models with tools like Ollama and LM Studio, and those with compatible Nvidia GPUs or newer Apple silicon, may find PAIR relevant.