PoCs a utility to get the totsl parameters count of a diffusers pipeline without hefty downloads.

get_params_diffusers_diffusers.py

from huggingface_hub import (
    hf_hub_download,
    list_files_info,
    parse_safetensors_file_metadata,
)
import json
import argparse


def get_files_info(args):
    files_info = list_files_info(args.repo_id, expand=True)
    files_info = list(files_info)
    return files_info


def get_model_index(args):
    model_index_path = hf_hub_download(
        repo_id=args.repo_id, filename="model_index.json"
    )

    with open(model_index_path) as f:
        model_index = json.load(f)

    return model_index


def get_params(args):
    model_index = get_model_index(args)
    components = [k for k in model_index.keys() if isinstance(model_index[k], list)]

    repo_files_info = get_files_info(args)

    total_params = 0
    for file in repo_files_info:
        if (
            len(file.path.split("/")[-1].split(".")) == 2
            and "safetensors" in file.path
            and any(k in file.path for k in components)
        ):
            total_params += parse_safetensors_file_metadata(
                repo_id=args.repo_id, filename=file.path
            ).parameter_count["F32"]

    return total_params


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Simple example of a training script.")
    parser.add_argument(
        "--repo_id",
        type=str,
        default="runwayml/stable-diffusion-v1-5",
        help="Repo id to count the params for.",
    )
    args = parser.parse_args()

    params = get_params(args)

    print(f"{args.repo_id} has {params / 1e9} billion parameters.")

model_size_poc_diffusers.ipynb

Some results:

runwayml/stable-diffusion-v1-5

python get_params_diffusers_diffusers.py

runwayml/stable-diffusion-v1-5 has 1.370216895 billion parameters.

stabilityai/stable-diffusion-xl-base-1.0

python get_params_diffusers_diffusers.py --repo_id=stabilityai/stable-diffusion-xl-base-1.0

stabilityai/stable-diffusion-xl-base-1.0 has 3.55249173 billion parameters.

stabilityai/stable-video-diffusion-img2vid-xt

python get_params_diffusers_diffusers.py --repo_id=stabilityai/stable-video-diffusion-img2vid-xt

stabilityai/stable-video-diffusion-img2vid-xt has 2.254442729 billion parameters.