Skip to content

Conversation

@linoytsaban
Copy link
Collaborator

add lora training script for https://huggingface.co/Tongyi-MAI/Z-Image

@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.

@linoytsaban linoytsaban marked this pull request as ready for review January 29, 2026 14:58
@linoytsaban linoytsaban requested a review from sayakpaul January 29, 2026 14:58
@linoytsaban
Copy link
Collaborator Author

@bot /style

@github-actions
Copy link
Contributor

github-actions bot commented Jan 29, 2026

Style bot fixed some files and pushed the changes.

@sayakpaul
Copy link
Member

Thanks for the script, do you have an example training run as well?

@linoytsaban
Copy link
Collaborator Author

linoytsaban commented Jan 29, 2026

running a testing train run with this dataset: https://huggingface.co/datasets/Norod78/Yarn-art-style
https://wandb.ai/linoy/dreambooth-z-image-lora/runs/yqd6pnzg?nw=nwuserlinoy
seems to be learning as expected
left/top: after 18 steps, right/bottom: after 1000 steps
a panda drinking matcha, yarn art style
media_images_validation_18_7b60e21cb9e7acc524a0
media_images_test_1000_d48f997c1155d8400f5d

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.

3 participants