Conversational AI Chatbot with Memory + RAG

A Streamlit chatbot with:

• Session-aware conversational memory
• Document-grounded RAG with ChromaDB + HuggingFace embeddings
• Streaming responses
• SQLite-backed session persistence
• Export chat to TXT/PDF

Run locally

1. Create a virtual environment and install dependencies:

   pip install -r requirements.txt

2. Start Ollama and pull a model (example):

   ollama pull mistral

3. Copy env file:

   cp .env.example .env

4. Launch app:

   streamlit run app.py

Notes

• Default model provider is Ollama via LangChain ChatOllama.
• Vector store is ChromaDB persisted in .chroma/.
• Sessions and messages are stored in chatbot.sqlite3.