Nvidia has released Personal AI Router, a free software tool that aggregates idle computers within a home network to function as a single local AI workstation. The open-source application scans for compatible machines, linking Nvidia GeForce RTX 20-series cards and newer alongside Apple M4 chips to handle inference tasks for local models. This setup allows users to run agentic workflows without sending data to external cloud servers. The software manages the distribution of computational load across connected devices, ensuring that available processing power is utilised for tasks like running Ollama or LM Studio locally.
The initiative addresses the growing demand for private, on-premise AI processing by turning scattered hardware into a unified resource. It lowers the barrier for individuals to run large language models without purchasing expensive dedicated servers. By pooling resources, users can access higher performance than a single device offers while maintaining data privacy. The tool simplifies the technical overhead of managing distributed computing environments for personal use.
- Supports Nvidia RTX 20-series and newer GPUs
- Compatible with Apple M4 chips and newer
- Runs entirely on local hardware without cloud dependency
