If you do not want to tweak parameters you can just run

    llama-server -m model.gguf

That's it