🤖 Little Agent Chatbot 🤖
- A lightweight local AI agent chatbot powered by Ollama and LangChain/LangGraph with RAG capabilities.
- Tested LLM: qwen3:1.7b; qwen3:4b; granite3.3:2b; llama3.2:3b; cogito:3b
- Dual provider support: a) local Ollama LLM b) remote Claude LLM (API-key required)
- Tested on low-budget hardware 8GB RAM.
- Included tools: real time weather, calculate, local documents RAG, local SQL database query and update.
🌟 Overview
Little Agent Chatbot is a simple yet powerful local AI assistant that runs entirely on your machine. Built for learning and experimentation, it combines the power of LLMs with advanced retrieval-augmented generation (RAG) to create an intelligent chatbot that can work with your personal documents and provide real-time information.
✨ Key Features
- 🏠 Fully Local: Runs completely on your machine - no data leaves your device
- 💰 Budget-Friendly: Works on low-resource hardware using efficient models.
- 🏠 Dual provider: a) local Ollama LLM b) remote Claude LLM (API-key required)
- 📚 RAG Integration: Upload and chat with your PDF documents using ChromaDB and Nomic embeddings
- 🌐 Real-Time Data: Get live weather information and perform calculations
- 🔧 Agent Framework: Extensible agent system built with Langchain
- 💻 Easy Interface: Clean web interface powered by Gradio
- 💻 Interface Options: Graphic web interface or classic text interface.
- 🔓 Open Source: MIT licensed and fully customizable
🚀 Tech Stack
- LLM Backend: Ollama with Qwen3 model (1.7B or 4B variants)
- Agent Framework: Langchain ecosystem
- Vector Database: ChromaDB for document embeddings
- Embeddings: Nomic-embed-text for semantic search
- Document Processing: PyPDF, Unstructured, TikToken
- Web Interface: Gradio
- Language: Python
- License: MIT
🎯 What Makes This Special?
- Privacy First: Your conversations and documents stay on your device
- Cost Effective: No API costs - runs on your hardware
- Educational: Perfect for learning about AI agents and RAG systems
- Extensible: Easy to modify and add new capabilities
- Lightweight: Designed to work on modest hardware setups
📖 Usage
Basic Chat
Simply type your questions and the AI will respond using the local Qwen3 model.
Document Upload
Upload PDF documents to enable RAG functionality. The chatbot will be able to answer questions based on your documents.
Agent Capabilities
- Weather Information: Get real-time weather data
- Calculations: Perform mathematical operations
- Document Q&A: Query your uploaded PDFs
🏗️ Architecture
User Input → Gradio Interface → Langchain Agent → Ollama/Qwen3 → Response
↓
RAG System (PDF Documents)
↓
External Tools (Weather, Calculator)
🔧 Configuration
Customize the chatbot by modifying:
- Model parameters in
config.py
- Agent tools and capabilities
- UI appearance and behavior
- RAG document processing settings
🤝 Contributing
Contributions are welcome! This project is designed for learning and experimentation. Feel free to:
- Add new agent capabilities
- Improve the UI/UX
- Enhance document processing
- Add new LLM models
- Fix bugs and improve performance
📝 License
This project is licensed under the MIT License - see the LICENSE file for details.
🎓 Educational Purpose
This project is perfect for:
- Learning about AI agents and RAG systems
- Understanding local LLM deployment
- Experimenting with Langchain framework
- Building privacy-focused AI applications
- Exploring document-based AI interactions
🔮 Future Enhancements
- Support for more document formats
- Additional agent tools and capabilities
- Multi-language support
- Voice interface integration
- Mobile-friendly interface
📞 Support
If you encounter issues or have questions:
- Open an issue on GitHub
- Check the documentation
- Review the example configurations
🌟 Star the Project
If you find this project helpful, please give it a star! It helps others discover the project and motivates continued development.
Made with ❤️ for the AI community
Happy chatting with your local AI agent! 🚀