quent_models.py

import os
from huggingface_hub import hf_hub_download
from typing import List

def QuantizeModel(model_name: str, lvl: str = 'q4_0'):
    """
    Function to quantize the model and save it in GGUF format.

    :param model_name: Path to the model (safetensors file)
    :param lvl: Quantization level (e.g., 'q4_0')
    """
    gguf_name = model_name.replace('.safetensors', f"_{lvl}.gguf")
    os.system(f'./bin/sd -M convert -m {model_name} -o {gguf_name} -v --type {lvl}')


def download_models(models: List[str]):
    """
    Download models from Hugging Face using their URLs.
    
    :param models: List of Hugging Face model URLs
    """
    for model_url in models:
        os.system(f'curl -L -O {model_url}')


# Example usage
if __name__ == "__main__":
    models_to_download = [
        "https://huggingface.co/stabilityai/stable-diffusion-3-medium/resolve/main/sd3_medium_incl_clips_t5xxlfp16.safetensors",
        "https://huggingface.co/CompVis/stable-diffusion-v-1-4-original/resolve/main/sd-v1-4.ckpt",
        "https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.safetensors",
        "https://huggingface.co/stabilityai/stable-diffusion-2-1/resolve/main/v2-1_768-nonema-pruned.safetensors"
    ]

    download_models(models_to_download)

    # Quantize models
    model_names = [
        'sd-v1-4.ckpt', 
        'v1-5-pruned-emaonly.safetensors', 
        'v2-1_768-nonema-pruned.safetensors',
        'sd3_medium_incl_clips_t5xxlfp16.safetensors'
    ]

    for model_name in model_names:
        QuantizeModel(model_name, lvl='q4_0')