| .. |
|
cldm
|
Fix controlnet not upcasting on models that have it enabled.
|
9 months ago |
|
extra_samplers
|
Cleanup uni_pc inpainting.
|
1 year ago |
|
k_diffusion
|
Fix some edge cases with samplers and arrays with a single sigma.
|
10 months ago |
|
ldm
|
Remove redundancy in mmdit.py (#3685)
|
9 months ago |
|
sd1_tokenizer
|
Initial commit.
|
2 years ago |
|
t2i_adapter
|
All resolutions now work with t2i adapter for SDXL.
|
2 years ago |
|
t5_tokenizer
|
SD3 Support.
|
9 months ago |
|
taesd
|
Fix taesd VAE in lowvram mode.
|
1 year ago |
|
checkpoint_pickle.py
|
Remove pytorch_lightning dependency.
|
2 years ago |
|
cli_args.py
|
Add debug options to force on and off attention upcasting.
|
9 months ago |
|
clip_config_bigg.json
|
This is unused but it should be 1280.
|
2 years ago |
|
clip_model.py
|
Always return unprojected pooled output for gligen.
|
1 year ago |
|
clip_vision.py
|
Replace prints with logging and add --verbose argument.
|
12 months ago |
|
clip_vision_config_g.json
|
Add support for clip g vision model to CLIPVisionLoader.
|
2 years ago |
|
clip_vision_config_h.json
|
Add support for unCLIP SD2.x models.
|
2 years ago |
|
clip_vision_config_vitl.json
|
Add support for unCLIP SD2.x models.
|
2 years ago |
|
conds.py
|
Fix bug with cosxl edit model.
|
9 months ago |
|
controlnet.py
|
Fix issue with controlnet models getting loaded multiple times.
|
11 months ago |
|
diffusers_convert.py
|
Fix saving text encoder in fp8.
|
11 months ago |
|
diffusers_load.py
|
Cleanup some unused imports.
|
1 year ago |
|
gligen.py
|
Fix gligen lowvram mode.
|
1 year ago |
|
latent_formats.py
|
SD3 Support.
|
9 months ago |
|
lora.py
|
dora_scale support for lora file.
|
11 months ago |
|
model_base.py
|
Don't auto convert clip and vae weights to fp16 when saving checkpoint.
|
9 months ago |
|
model_detection.py
|
SD3 Support.
|
9 months ago |
|
model_management.py
|
Load the SD3 T5xxl model in the same dtype stored in the checkpoint.
|
9 months ago |
|
model_patcher.py
|
Remove useless code.
|
9 months ago |
|
model_sampling.py
|
SD3 Support.
|
9 months ago |
|
ops.py
|
Fix OSX latent2rgb previews.
|
9 months ago |
|
options.py
|
Only parse command line args when main.py is called.
|
1 year ago |
|
sample.py
|
Reshape the empty latent image to the right amount of channels if needed.
|
9 months ago |
|
sampler_helpers.py
|
Refactor sampler code for more advanced sampler nodes part 2.
|
11 months ago |
|
samplers.py
|
Support sampling non 2D latents.
|
9 months ago |
|
sd.py
|
Single and dual clip loader nodes support SD3.
|
9 months ago |
|
sd1_clip.py
|
Load the SD3 T5xxl model in the same dtype stored in the checkpoint.
|
9 months ago |
|
sd1_clip_config.json
|
Initial commit.
|
2 years ago |
|
sd2_clip.py
|
Remove some unused imports.
|
9 months ago |
|
sd2_clip_config.json
|
Move text_projection to base clip model.
|
2 years ago |
|
sd3_clip.py
|
Single and dual clip loader nodes support SD3.
|
9 months ago |
|
sdxl_clip.py
|
Load the SD3 T5xxl model in the same dtype stored in the checkpoint.
|
9 months ago |
|
supported_models.py
|
Single and dual clip loader nodes support SD3.
|
9 months ago |
|
supported_models_base.py
|
Fix issue with sampling_settings persisting across models.
|
11 months ago |
|
t5.py
|
SD3 Support.
|
9 months ago |
|
t5_config_base.json
|
SD3 Support.
|
9 months ago |
|
t5_config_xxl.json
|
SD3 Support.
|
9 months ago |
|
types.py
|
Fix UnetParams annotation typo (#3589)
|
9 months ago |
|
utils.py
|
Reshape the empty latent image to the right amount of channels if needed.
|
9 months ago |
