now how about you implement for the user to actually test it with their LLM model locally running on local lm-studios api server (localhost:1234)

initial commit