Hi Community!
I was trying to use stable diffusion 2 to run the paper below.
Unfortunately, the model is not available. Is it expected?
https://huggingface.co/stabilityai/stable-diffusion-2
Thanks for your help!
Hi Community!
I was trying to use stable diffusion 2 to run the paper below.
Unfortunately, the model is not available. Is it expected?
https://huggingface.co/stabilityai/stable-diffusion-2
Thanks for your help!
Hmm… Same here for both stabilityai/stable-diffusion-2-1-base and stabilityai/stable-diffusion-2-1…
Seems same there…?
Same goes for stabilityai/stable-diffusion-2-inpainting
Any idea what this 404 error means? Does it mean that the model has been retracted?
Does it mean that the model has been retracted?
Yeah. Or the model may have been made private, but from an external perspective, it makes no difference (just 404 not found). The reason is unclear, but since Stable Diffusion experienced the same disappearance issue with SD 1.5 before, this occurrence itself is not surprising.
Well this is hugely disappointing. I was in the middle of working on a project with this. Is there anything that can be done? Someone to contact?
Is there anything that can be done?
While the number of times you can use it per day is limited, backing up remaining model weights with Repo duplicator should be the top priority. If that doesn’t work, simply download them locally (about 101GB).
# pip install -U huggingface_hub[hf_xet]
import shutil
from pathlib import Path
from typing import Optional
from huggingface_hub import snapshot_download
def dl(repo_id: str) -> Optional[str]:
local_dir = str(Path("./", f"{repo_id.split('/')[0]}-{repo_id.split('/')[-1]}"))
cache_dir = str(Path("./", f"{repo_id.split('/')[0]}-{repo_id.split('/')[-1]}-cache"))
try:
if Path(local_dir).exists():
print(f"{local_dir} alrealy esxists. Skipping {repo_id}.")
return None
if Path(cache_dir).exists():
print(f"{cache_dir} alrealy esxists. Skipping {repo_id}.")
return None
print(f"Downloading {repo_id} to {local_dir}...")
snapshot_download(repo_id=repo_id, local_dir=local_dir, cache_dir=cache_dir)
local_cache_path = str(Path(local_dir, ".cache"))
if Path(local_cache_path).exists(): shutil.rmtree(local_cache_path)
if Path(cache_dir).exists(): shutil.rmtree(cache_dir)
print(f"Downloaded {repo_id} to {local_dir}...")
except Exception as e:
print(f"Error: {e}")
if Path(local_dir).exists(): shutil.rmtree(local_dir)
if Path(cache_dir).exists(): shutil.rmtree(cache_dir)
return None
return local_dir
if __name__ == "__main__":
dl("jyoung105/stable-diffusion-v2-1")
dl("Manojb/stable-diffusion-2-base")
dl("Manojb/stable-diffusion-2-1-base")
dl("alwold/stable-diffusion-2-inpainting")
Someone to contact?
For cases like this, I’m not sure if even HF support can help…
But for now, Hugging Face’s support is: [email protected]
Verify helpful information John6666! Thanks a lot!