Add partial SDXL model #61
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
coryMosaicML
approved these changes
Aug 18, 2023
Landanjs
reviewed
Aug 18, 2023
There was a problem hiding this comment.
Looks great! The only major adjustment is the NotImplementedError for the pretrained arg. I also had a question around which VAE we should prefer if we can run it on bf16.
For the future, there should be a future PR that adds the other modeling features (in addition to the conditioning PR), so we can run a pre-trained SDXL. Sorry for going back-and-forth on whether to fully implement SDXL
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This PR adds
diffusion.models.models.stable_diffusion_xl, which incorporates a few ideas from SDXL. Namely:Currently we're using the
madebyollin/sdxl-vae-fp16-fixVAE checkpoint rather than the officialstabilityai/stable-diffusion-xl-base-1.0VAE checkpoint. The stabilityai checkpoint leads to NaNs when training with fp16. More on that issue here.