Skip to content

Conversation

@mbalabanski
Copy link
Contributor

@mbalabanski mbalabanski commented Dec 26, 2025

What does this PR do?

This PR changes the checkpoint_mapping_fn of the single_file loader of QwenImageTransformer2DModel to maintain a consistent API with the other loaders. Currently, the checkpoint mapping function is set to be a lambda which takes in an argument x and returns this same argument.

This PR fixes an issue caused by this lambda to use more consistent arguments as with the other checkpoint_to_diffusers functions.

Fixes #12891

Before submitting

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@HuggingFaceDocBuilderDev

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@DN6 DN6 merged commit 208cda8 into huggingface:main Jan 2, 2026
10 of 11 checks passed
@DN6
Copy link
Collaborator

DN6 commented Jan 2, 2026

Thanks @mbalabanski 👍🏽

Aki-07 pushed a commit to bconstantine/diffusers that referenced this pull request Jan 4, 2026
… consistent with other loader APIs (huggingface#12894)

fix Qwen single file loading to be consistent with other loader API
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.

not support load Qwen-Image-Edit-2511 gguf or fp8 safetensors?

3 participants