LLM Applications with Haystack and Streamlit

The aim of this repository is to demonstrate how to build LLM applications using Haystack and Streamlit.

How to start

• It is assumed that an Ollama instance is running on the local machine.
• The following environmental variables are needed.
  • HF_TOKEN: Token of Huggingface.

streamlit run app.py

TODOs

• Tests
• docker compose
• CI/CD pipeline on AWS
• More tools such as simple agents