AI News Feed
Market watch
LG UltraFine Evo 32U990A-S Hits Record-Low $1,099 on Amazon, TechRadar ReportsNewegg Labor Day Deal: 32GB Kingston Fury Beast DDR4-3200 Kit Nearly Half Price at $179.99 with CodeGalaxy Z Fold 8 inner screen corners reported with excessive giveHuawei unveils Mate XT 2 tri-fold with privacy-screen edition as Apple prepares folding iPhoneWall Street Analysts Recommend Nvidia, Uber and Marvell for Long-Term GrowthLenovo, AMD and Huawei Put Local Large-Model AI at Center of New Devices
LG UltraFine Evo 32U990A-S Hits Record-Low $1,099 on Amazon, TechRadar ReportsNewegg Labor Day Deal: 32GB Kingston Fury Beast DDR4-3200 Kit Nearly Half Price at $179.99 with CodeGalaxy Z Fold 8 inner screen corners reported with excessive giveHuawei unveils Mate XT 2 tri-fold with privacy-screen edition as Apple prepares folding iPhoneWall Street Analysts Recommend Nvidia, Uber and Marvell for Long-Term GrowthLenovo, AMD and Huawei Put Local Large-Model AI at Center of New Devices
Products & Applications

Nvidia launches free PAIR tool that pools idle home PCs for local AI computing

Nvidia unveiled PAIR, a free tool that pools idle home PCs into one local AI compute cluster for agentic tasks.

AI News Feed · 2026-09-04 10:00

Despite its name, PAIR is not hardware. The software, available as a beta for Windows, Linux and macOS, discovers compatible systems on a local network and readies them for processing. PAIR works with Nvidia GeForce RTX 20-series GPUs and newer, RTX PRO workstation GPUs, DGX Spark systems and Apple Macs with M4 or newer chips, according to Nvidia.

When a local AI agent is given a task, PAIR identifies the subtasks that need to be done and distributes them across available GPU nodes so they do not all compete for the same machine. Completed results are returned to the main node. The system is elastic: if a user starts gaming or otherwise uses a machine in the middle of a subtask, PAIR moves that work to another idle node or back to the main node when no other capacity is available.

Nvidia product manager Seth Schneider said in a briefing that PAIR is designed for a realistic household with about one MacBook or Windows laptop and a gaming PC. He also described a family with several high-end systems, estimating that setup has roughly 165 teraflops of underutilized compute. “It’s truly a treasure trove of free tokens just sitting in homes today,” Schneider said.

Enrolling machines requires installing PAIR and the client will find PCs through mDNS or IP addresses, then help initiate model downloads on each node. Nodes do not need identical models; PAIR checks what is available on each PC and distributes work accordingly. The tool supports AI front-ends such as LM Studio and Ollama through automatically created proxies. Nvidia said device pairing is protected by a six-digit code and mTLS, creating an encrypted, mutually trusted channel.

PAIR is now available in beta for macOS, Windows and Linux. Nvidia also announced that Perplexity Portable Computer, Hermes Agent and OpenClaw, three agentic AI applications, will soon offer simplified local setup with Nvidia GPUs on Windows, cutting the manual configuration to a few clicks.