RagBot is an advanced Retrieval-Augmented Generation (RAG) chatbot that integrates Ollama and LangChain to provide intelligent, data-driven conversations. It can work with or without internal company data, making it a versatile AI assistant. The chatbot is accessible via a web browser and runs on a Flask backend. 🚀
- Ollama Integration 🏗️: Easily switch between different AI models.
- Retrieval-Augmented Generation (RAG) 📚: Uses pre-existing company data for enhanced responses.
- Web Deployment 🌐: Simple HTML/JS frontend for easy interaction and modify the chatbot UI with ease.
- Flask Support 🔥: For backend framework.
.
├── ChatBot.py # Core chatbot logic (without company data)
├── RagBot.py # RAG-enhanced chatbot (with company data)
├── Internal_Data.txt # Example company data (FAKE DATA)
├── app.py # Flask server
├── requirements.txt # Python dependencies
├── set_environment.sh # Script to set up the environment
├── templates
│ └── index.html # Web UI for chatbot
├── img # Images for documentation
│ ├── ChatBot.png # Basic chatbot example
│ ├── RagBot.png # RAG-powered chatbot example
│ ├── WebBot1.png # Web UI example 1
│ ├── WebBot2.png # Web UI example 2
To create a virtual environment and install dependencies, run:
bash set_environment.shOllama is required for running AI models locally. Install it using:
curl -fsSL https://ollama.ai/install.sh | shFor Windows users, download from Ollama's official site, even if there is issue in any other OS, please refer to their website.
Pull the Llama 3 model into Ollama:
ollama pull llama3Manually start and serve the model, using the following command:
ollama serveSimply start the chatbot server:
python app.pyThen, open your browser and visit:
http://127.0.0.1:8000/
There is a small chat button, on the bottom right. Click on it and begin chatting with the RagBot 💬.
If you want to test the bots, without the webbrowser, you can run either python ChatBot.py or python RagBot.py in your terminal to chat with the bot without and with RAG enabled respectively.
LangChain provides an easy way to build AI-powered applications by connecting models, data sources, and logic. It allows RagBot to:
- Retrieve relevant data efficiently 🗄️
- Use memory to maintain conversational context 🧠
- Seamlessly integrate with multiple AI models ⚙️
Ollama enables running large language models (LLMs) locally with ease. Here are some of the best options:
| Model Name | Use Case 🛠️ |
|---|---|
| LLaMA 3 🦙 | General-purpose chatbot, reasoning, and Q&A |
| Mistral 7B 🌪️ | Lightweight, optimized for fast responses |
| Gemini Nano ✨ | Multimodal AI (text + images) for creative tasks |
| Code Llama 💻 | Code generation and programming assistance |
| Phi-2 🧠 | Small, efficient model for lightweight apps |
| Custom Fine-Tuned Models 🔬 | Tailored to your business data and needs |
A large language model by Meta, focused on efficient, open-source NLP tasks like text generation and question-answering.
A library by Meta for fast, efficient similarity search and clustering of high-dimensional vectors, commonly used in search and recommendation systems.
A lightweight model for generating sentence embeddings, optimized for fast and efficient semantic text comparison.
In our Retrieval-Augmented Generation (RAG) approach, we use the following:
- Embedding Model:
"sentence-transformers/all-MiniLM-L6-v2" - RAG Pipeline: Uses
Internal_Data.txtfor contextual responses - LLM: Runs on Ollama with LLaMA 3
Want to improve RagBot? Follow these steps:
- Fork the repo 🍴
- Create a new branch 🛠️ (
git checkout -b feature-name) - Make your changes 💡
- Submit a Pull Request (PR) 🔥
📢 All data used in this project is FAKE and for demonstration purposes only.
If you have any questions or run into any issues, feel free to reach out to me on LinkedIn. I'm here to help! 😊