RAG Chatbot - FastAPI + Next.js

An interactive chatbot powered by Retrieval-Augmented Generation (RAG), built with FastAPI backend and a Next.js frontend. The system retrieves relevant documents from a local vector database (ChromaDB) and generates context-aware answers through a custom RAG pipeline. It uses local open-source llama3-1 model, so it can answer some general knowledge questions without needing any external documents.

The project is split into two independent parts:

• backend/ - FastAPI, RAG pipeline, ChromaDB, ingestion scripts
• frontend/ - Next.js chat interface

---

Features

• RAG pipeline for document retrieval
• FastAPI backend with clean modular architecture
• Next.js chat UI with loading animation and message history
• Real-time communication between frontend and backend
• Embeddings + vector search via ChromaDB
• CORS-enabled for safe frontend <-> backend communication
• Easily extendable for new models, data sources or UI features

---

Getting started

1. Clone the repo

# In terminal
git clone https://github.com/bjukic2/Python-RAG-Chatbot.git

# Navigate to the project folder
cd Python-RAG-Chatbot

2. Install dependencies

# Install dependencies for backend
cd backend/
source venv/bin/activate
pip install -r requirements.txt

# Return to previous folder
cd ..

# Install dependencies for frontend
cd frontend/
npm install

# Return to previous folder
cd ..

3. Installing Llama using Ollama

curl -fsSL https://ollama.com/install.sh | sudo sh
ollama pull llama3.1
ollama run llama3.1

4. Run the project

# Open two terminals

# In terminal one run backend
cd backend/
uvicorn main:app --reload

# In terminal two run frontend
cd frontend/
npm run dev

5. Use the chatbot

Open http://localhost:3000 and type your question in the text field and press "Send"

---

Author

Made by Bruno Jukić [https://github.com/bjukic2]