NVIDIA PAIR, now in beta, routes local AI inference requests across NVIDIA DGX Spark, Windows RTX systems, and macOS devices through a single endpoint. The software discovers compatible machines on a home network and distributes workloads across them without combining devices into one virtual GPU. At launch, PAIR supports Ollama and LM Studio backends on Windows, Linux, and macOS. Prompts, files, and agent context stay on the user's local network, and no internet connection is required for operation.
No score is assigned. Sources and their independence are shown in the citation chain below.