datvodinh/rag-chatbot

Chat with multiple PDFs locally

🤖 Chat with multiple PDFs locally

📖 Table of Contents

• Feature
• Idea
• Setup
  • Kaggle
  • Local
    • Clone
    • Install
    • Run
• Todo

⭐️ Key Features

• Easy to run on Local or Kaggle (new)
• Using any model from Huggingface and Ollama
• Process multiple PDF inputs.
• Chat with multiples languages (Coming soon).
• Simple UI with Gradio.

💡 Idea (Experiment)

💻 Setup

1. Kaggle (Recommended)

• Import notebooks/kaggle.ipynb to Kaggle
• Replace <YOUR_NGROK_TOKEN> with your tokens.

2. Local

2.1. Clone project

git clone https://github.com/datvodinh/rag-chatbot.git
cd rag-chatbot

2.2 Install

2.2.1 Docker

docker compose up --build

2.2.2 Using script (Ollama, Ngrok, python package)

source ./scripts/install_extra.sh

2.2.3 Install manually

1. Ollama

• MacOS, Window: Download

• Linux

curl -fsSL https://ollama.com/install.sh | sh

2. Ngrok

• Macos

brew install ngrok/ngrok/ngrok

• Linux

curl -s https://ngrok-agent.s3.amazonaws.com/ngrok.asc \
| sudo tee /etc/apt/trusted.gpg.d/ngrok.asc >/dev/null \
&& echo "deb https://ngrok-agent.s3.amazonaws.com buster main" \
| sudo tee /etc/apt/sources.list.d/ngrok.list \
&& sudo apt update \
&& sudo apt install ngrok

3. Install rag_chatbot Package

source ./scripts/install.sh

2.3 Run

source ./scripts/run.sh

or

python -m rag_chatbot --host localhost

• Using Ngrok

source ./scripts/run.sh --ngrok

3. Go to: http://0.0.0.0:7860/ or Ngrok link after setup completed

🎯 Todo

• Add evaluation.
• Support better Embedding Model for Vietnamese and other languages.
• Knowledge Graph (for Structure Data).
• Better Document Processing.
• MLX model.

🌟 Star History

rag-chatbot's Issues

ngrok requires sudo priviledges to install - not suitable for local installation where users can't use sudo

not suitable for local installation where users can't use sudo

telegram bot integration

it would be perfect to integrate with a telegram bot (such as ollama-telegram-bot or something similar).