GitLab

* Let tokenizers return weights to be stored in the saved checkpoint.

* Basic hunyuan dit implementation.

* Fix some resolutions not working.

* Support hydit checkpoint save.

* Init with right dtype.

* Switch to optimized attention in pooler.

* Fix black images on hunyuan dit.

Set the default AuraFlow shift value to 1.73 (sqrt(3)).

You can use the CLIPLoader to use the t5xxl only or the DualCLIPLoader to
use CLIP-L and CLIP-G only for sd3.

This is the code to load the model and inference it with only a text
prompt. This commit does not contain the nodes to properly use it with an
image input.

This supports both the original SD1 instructpix2pix model and the
diffusers SDXL one.

Use the UNET Loader node to load the unet file to use them.

Fix issue with not loading the SSD1B clip correctly.

This is an old model.

Load the checkpoint like a regular one and use the new
SD_4XUpscale_Conditioning node.

To use it use the ImageOnlyCheckpointLoader to load the checkpoint and
the new Stable_Zero123 node.