Everything need
Create GGUF quantized model from a Hugging Face repo
Calculate VRAM needed to run LLMs on your GPU
