Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

(SD) Bump punet revision to d30d6ff and enable punet test. #774

Open
wants to merge 41 commits into
base: main
Choose a base branch
from

Conversation

monorimet
Copy link
Contributor

The SHA may need to be updated once the fused qkv model is merged in the amd-shark/sdxl-quant-models repo.

elif precision in ["fp16", "fp32"]:
repo_id = hf_model_name
subfolder = "unet"
revision = "76d28af79639c28a79fa5c6c6468febd3490a37e"
revision = "d30d6ff79abb584bf2addc7866738df5242f315a"
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Shouldn't need to be updated. I'll just merge to main but the commit should stay the same.

We can use tags on hf and may want to do that at some point.

monorimet and others added 15 commits July 22, 2024 11:58
…l with all the new changes. (#785)

This PR makes the necessary changes so that you don't have to run the
quantization docker to run the normal harness. With stricter checks for
the quant paths and adding a special switch to the save_external_weights
so that it is compliant with this change that we need for the docker
generated weights
iree-org/iree-turbine@cd916ec.
Sync flags to sdxl-scripts repo
Currently using a link, but Nithin will be pushing the fix to use a file
name asap
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants