llame llame generates commit messages for you with the help of LLMs. For now only the llama-server hosted models (from llama.cpp project) are supported.