OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Airgapped Offline RAG

Airgapped Offline RAG is a Python application that processes uploaded PDFs and generates answers to users’ questions through a Streamlit interface, using retrieval, vector storage and selectable language models.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Airgapped Offline RAG — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a local interface for asking questions about PDFs and can prepare the Python environment, model files and first-run embedding dependency.

Who it is for

  • People who want to query PDF collections through a locally run application.
  • Developers exploring a self-hosted RAG prototype built with LangChain, ChromaDB and Streamlit.

Documented uses

  • Upload PDFs and ask questions grounded in their contents.
  • Select a supported model for document-based question answering.
  • Adjust model paths and chunk sizes through config.yaml.
  • Run the application locally or invoke its documented Docker targets.
Before you choose

Requirements & limits

The project is public on GitHub and licensed under GPL-3.0. Its README documents local and Docker run targets.

  • Python 3.9 and project dependency setup are required for the documented local path.
  • GGUF model files must be downloaded and placed in the models/ directory.
  • The Qdrant Sentence Transformer model downloads on first run, so the README recommends initial internet access.
  • The described document upload workflow is for PDFs; additional document formats are listed as future features.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up and run the local PDF question-answering app

Prepare the documented Python environment and launch the app locally for PDF-based questions.

Before you start

  • Python 3.9
  • Project dependencies installed with make setup
  • A downloaded GGUF model in the models/ directory
  • Internet access on the initial run to download the Qdrant Sentence Transformer model

Steps

  1. Create the project environment and install dependencies with make setup.

  2. Download a supported GGUF model and place it in the models/ directory.

  3. Run make run to start the application.

  4. Upload PDF documents, select a model, enter a question, and request an answer.

Expected outcome

A locally launched interface for generating answers based on uploaded PDF content.

Guide limitations

  • The setup instructions recommend internet access initially because the Sentence Transformer model downloads on first run.
  • The documented upload workflow supports PDFs; non-PDF and image document support is listed as future work.
Documentation 1 ↗
Source-based practical guide

Build and run the Docker version

Use the repository's documented Docker targets to build and launch the application.

Before you start

  • The repository's dependencies and model setup prepared as described in the README
  • Docker

Steps

  1. Run make docker-build.

  2. Run make docker-run.

Expected outcome

The README's Docker build and run targets are invoked for the application.

Guide limitations

  • The README does not provide further Docker configuration or deployment instructions.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
vincentkoc/airgapped-offfline-rag · Official homepage title · retained in its source language
How we review sourcesSuggest a correction