When running local LLMs, many people choose Ollama first. This is because it is easy to set up and makes model management simple. On the other hand, Ollama uses an engine internally based on llama.cpp ...