EduRAG

A fully local Retrieval-Augmented Generation (RAG) chatbot and automatic quiz generator with no telemetry and usage costs

✨ Features

• Conversational chatbot with contextual memory powered by your own documents
• Automatic quiz generator based on the uploaded content
• 100% local AI using the llama3.2:3b model running via Ollama, no external API calls
• RAG (Retrieval-Augmented Generation) with nomic-embed-text embeddings
• Local vector database (Milvus Standalone) for efficient semantic search
• Object storage with MinIO
• Complete privacy, all data stays on your machine

Prerequisites

• Docker and Docker Compose
• Ollama installed
• ~8 GB of available RAM
• ~5 GB of disk space

Installation & Setup

1. Install Ollama and pull the models

Download and install Ollama, then run:

ollama pull llama3.2:latest
ollama pull nomic-embed-text

2. Clone the repository

git clone https://github.com/rubenzu03/RAG_chatbot.git
cd RAG_chatbot

3. Configure environment variables and edit the .env file with your values

cp .env.example .env

4. Start the containers

docker compose up -d

5. Set up MinIO

1. Open http://localhost:9001 in your browser
2. Log in with the credentials from your .env file
3. Create a bucket named ragchatbot
4. Upload the documents you want to use as the knowledge base (PDF, TXT, DOCX, MD...)

6. Re-index the documents

After uploading your files, restart the containers so the RAG pipeline processes and indexes the content:

docker compose down
docker compose up -d

7. Access the application

Open http://localhost:5173

Built with Llama