GGUF

The standard file format for quantized models in the llama.cpp/Ollama ecosystem — the reason a 7B model runs on a laptop.