Portable-RAG: A network-independent, personal computer-based AI tool for teaching and research

The growing need to ensure data confidentiality and reduce operational costs has driven the development of AI solutions that do not rely on cloud infrastructures. This software introduces a portable Retrieval-Augmented Generation (RAG) system, designed to operate entirely offline on a laptop, without requiring any network connection. The solution enables researchers to query and generate content based on local documents, maintaining full control over sensitive data and complying with privacy constraints. The proposed architecture integrates a semantic search engine and an optimized generative model for consumer-grade hardware, ensuring high performance and ease of use. This approach eliminates expenses associated with cloud services and minimizes data exposure risks, offering a secure and cost-effective alternative for research activities in academic and industrial contexts.