Local retrieval lab

RAG Lab

Build retrieval-augmented generation around your own documents while keeping the model stack on-device. The Windows package bundles the offline runtime, models, and local storage for a fully local workflow.

Open Online Version Download RAG-Lab-Portable.7z Back to Edgy AI

What ships

Local-first retrieval without a cloud dependency

RAG Lab is built around local LLMs and local retrieval. The goal is to answer from your own files with a model stack that stays close to the machine, not a remote service.

Document ingest Split files into chunks and push them into vector storage.

Source-backed answers Surface passages that explain where the answer came from.

Portable bundle The Windows build carries the offline runtime and models.