allow loading of sd models from safetensors without online lookups using local config files #5019
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Fixes outstanding critical request that diffusers can use
from_single_file
method to load safetensors filewithout needing to download yaml config file from github
This allows loading of safetensors to be done while offline.
Without this, any attempt to load safetensors model requires connection to github
(e.g. https://raw.githubusercontent.com/CompVis/stable-diffusion/main/configs/stable-diffusion/v1-inference.yaml)
Note that
download_from_original_stable_diffusion_ckpt
already has definedmodel_files
param and its fully documented,but it was never actually implemented. this PR completes the implementation in existing code.
Tested in SD.Next:
Before submitting
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
@patrickvonplaten and @sayakpaul