LLM (Large Language Model) based chatbot using RAG (Retrieval-augmented generation) and langchain. Chatbot has access to trade register documents related to Ticos Systems company and is able to answer any question related to this company not only giving answer but also referencing the source document.
as language model OpenAI GPT-4o was used to run chat in browser streamlit library was used
-
before run install needed libraries as below:
pip install -r requirements.txt -
create .env file with OpenAI token environment variable
OPENAI_API_KEY="your_token_from_openai" -
run script with command:
streamlit run index.py
To run it for your purpose you can replace source documents in /data directory with your files. Prompt stored in ChatPrompt.py also can be modified to fit your needs.
This project was originally based on https://github.com/shashankdeshpande/langchain-chatbot (currently updated with modern LangChain syntax and not much in common).