To test, project

Run

docker compose up --build -d

This will set up container containing llama3.2 this can be modified in ollama/entrypoint.sh use small models,