Local RAG: Models & Workflows
Chat with your own documents. Index files into a vector collection, retrieve the passages that match a question, and have an agent answer from them — on one canvas, on your own keys or fully local.
Models for rag
Open-weight embedding model for local vector indexing.
Fast, low-cost answer generation over retrieved passages.
Strong grounded answers with citations from long context.
Templates that wire rag
Frequently asked questions
What is RAG?
Retrieval-augmented generation searches your own documents for the passages relevant to a question, then passes those passages to a model so it answers from your material instead of guessing from training data.
Where does my document index live?
In NodeTool's own vector store (sqlite-vec), on your machine by default. Nothing has to leave your computer — embeddings and retrieval can run entirely on local models.
How do I run RAG in NodeTool?
Open the Chat With Your Documents template below, index a folder of files into a collection, and connect the query and agent nodes — no separate vector database to stand up.
Do rag on your own machine
Free, open source, and yours to run. Download Studio, pick a model, and wire it into a workflow with your own keys.