NVIDIA Personal AI Router - Turn your PCs and Macs into a private AI cluster

NVIDIA PAIR turns the computers on your local network into a personal AI inference cluster. Connect RTX PCs, DGX Spark systems, and Macs, then run local AI apps and agents through one endpoint while PAIR automatically routes requests to available machines. It works with Ollama and LM Studio across Windows, Linux, and macOS — while keeping prompts, files, and agent context on your network.

Add a comment

Replies

Best
If you have a gaming PC, a MacBook, and another computer sitting around the house, NVIDIA just found a new use for all that idle compute. PAIR connects them into a personal AI cluster and automatically routes local inference requests to whichever machine has capacity. Your AI apps see one endpoint, while PAIR handles the RTX PCs, Macs, or DGX Spark machines behind it. Everything can stay on your local network, and it already works with Ollama and LM Studio. Feels especially timely as running multiple local AI agents starts becoming a real use case.