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,