# mrdbourke/simple-local-rag

Build a RAG (Retrieval Augmented Generation) pipeline from scratch and have it all run locally.

Repository: https://github.com/mrdbourke/simple-local-rag
Canonical: https://ross.abutalabs.com/products/simple-local-rag
Language: Jupyter Notebook
License Family: other
Last push: 2024-05-25T11:26:31+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 64
- inputs: {"age_days": 903, "days_push": 830, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1011, forks 303 (observed 2026-08-28T04:03:13.113041+00:00)

## What it is
A tutorial repository by mrdbourke teaching how to build a retrieval-augmented generation (RAG) pipeline from scratch that runs entirely locally on an NVIDIA GPU. It walks through PDF ingestion, chunking, embedding with sentence transformers, semantic search, and LLM-based answer generation, culminating in a 'chat with PDF' style app called NutriChat.

## Use cases
- build a rag pipeline from scratch
- chat with pdf documents locally
- learn how retrieval augmented generation works
- run an llm question answering system on my own gpu
- embed pdf text and search it semantically
- follow a hands-on rag tutorial with jupyter notebooks

## When to choose
- you want to learn RAG concepts by building a complete pipeline yourself
- you have an NVIDIA GPU (5GB+ VRAM) or Google Colab access and want everything running locally without paid APIs
- you prefer open-source tools like PyTorch, sentence-transformers, and open LLMs over proprietary services

## When to avoid
- you need a production-ready, maintained RAG framework rather than educational notebook code
- you have no NVIDIA GPU and cannot use Colab
- you need a polished application with setup docs and support - the repo has incomplete setup instructions and no license

## Facets
- artifact type: learning-resource
- maturity: active
- function: rag, llm-inference, pdf, machine-learning, nlp
- domain: large-language-models, tutorials, artificial-intelligence, pdf
- platform: python, cross-platform
- tags: tutorial, jupyter-notebook, chat-with-pdf, local-llm, sentence-transformers, pytorch, open-source-llm, retrieval-augmented-generation, gpu

## Member repositories
- mrdbourke/simple-local-rag (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:13.113041+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:11:51.741944+00:00, confidence not recorded.
  - readme: https://github.com/mrdbourke/simple-local-rag (fetched 2026-08-28T04:03:13.113041+00:00, sha 4e8580ba0975)
- Data as of 2026-08-30T08:39:29.467469+00:00.