AI BriefingNVIDIA Open-Sources Local AI Inference Router 'PAIR'
Key point
NVIDIA has released 'PAIR', an open-source tool that connects multiple local computers to distribute AI inference requests.
Details
NVIDIA has released Personal AI Router (PAIR), an open-source tool that automatically distributes local AI inference requests by connecting multiple computers within a home. This tool clusters nodes running compatible inference engines such as Ollama and LM Studio, routing independent requests to the most suitable node based on engine availability, model availability, and current workload.
Key Features and Operations
- Automatic Routing: Distributes requests across nodes based on engine and model availability and workload.
- Engine Management: Allows automatic download and configuration of Ollama or LM Studio within PAIR, and supports integration with existing installed engines.
- Supported Environments: Supports Windows 11, Linux, and macOS (x64/arm64), with cross-OS node pairing. Windows ARM is experimental.
- Installation Method: Provides signed installers (.exe, .deb, .dmg) and can be run via a terminal interface if no desktop environment is available.
License and Security
- License: Follows the Apache License 2.0 and is available in the GitHub repository (NVIDIA/Personal-AI-Router).
- Security: Uses PIN-based trust bootstrap and LAN discovery; reviewing the security guidelines (SECURITY.md) is required before deployment on untrusted networks.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.