How to Run GGUF Models Locally: Ollama, llama.cpp & vLLM

How to run GGUF models locally: one-line Ollama pulls, llama.cpp straight off a Hugging Face URL, and how to pick the right quant for your VRAM.