Add support for lumina2 #10642
Open
Add support for lumina2 #10642
+1,475
−3
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
yiyixuxu
reviewed
Jan 24, 2025
yiyixuxu
reviewed
Jan 24, 2025
yiyixuxu
reviewed
Jan 29, 2025
| cap_freqs_cis[i, :cap_len] = freqs_cis[i, :cap_len] | ||
| img_freqs_cis[i, :img_len] = freqs_cis[i, cap_len:cap_len+img_len] | ||
|
|
||
| for layer in self.context_refiner: |
There was a problem hiding this comment.
I think context_refiner and noise_refiner should be moved inside foward for better readbility
does it make sense to wrap the code to create freqs_cis inside a class like this
There was a problem hiding this comment.
Thanks for the suggestion! I have moved refiner into forward() for better readbility. As for the freqs_cis, we already have a class Lumina2PosEmbed.
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Outdated suggestions cannot be applied.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.
What does this PR do?
This PR will add the official Lumina-Image 2 to the diffusers. Lumina-Image 2.0 is the latest model in the Lumina family and will be released very soon (https://huggingface.co/Alpha-VLLM/Lumina-Image-2.0). It is a 2B parameter Diffusion Transformer that significantly improves instruction-following and generates higher-quality, more diverse images. Our paper will be released soon, and we have finished the diffuser pipeline for Lumina-Image 2.0.
Core library: