comfyanonymous ba9095e5bd Automatically use fp8 for diffusion model weights if: ...

Checkpoint contains weights in fp8.

There isn't enough memory to load the diffusion model in GPU vram.

2024-08-03 13:45:19 -04:00

..

cldm

"auto" type is only relevant to the SetUnionControlNetType node.

2024-07-22 11:30:38 -04:00

extra_samplers

Cleanup uni_pc inpainting.

2024-02-23 02:39:35 -05:00

k_diffusion

Switch to the real cfg++ method in the samplers.

2024-06-29 11:59:48 -04:00

ldm

Hack to make all resolutions work on Flux models.

2024-08-01 21:39:18 -04:00

sd1_tokenizer

Initial commit.

2023-01-16 22:37:14 -05:00

t2i_adapter

Controlnet refactor.

2024-06-27 18:43:11 -04:00

taesd

Improvements to the TAESD3 implementation.

2024-06-16 02:04:24 -04:00

text_encoders

Fix clip_g/clip_l mixup (#4168)

2024-08-01 21:40:56 -04:00

checkpoint_pickle.py

Remove pytorch_lightning dependency.

2023-06-13 10:11:33 -04:00

cli_args.py

Make default hash lib configurable without code changes via CLI argument (#3947)

2024-07-16 18:27:09 -04:00

clip_config_bigg.json

Fix potential issue with non clip text embeddings.

2024-07-30 14:41:13 -04:00

clip_model.py

Lower CLIP memory usage by a bit.

2024-07-31 01:32:35 -04:00

clip_vision_config_g.json

Add support for clip g vision model to CLIPVisionLoader.

2023-08-18 11:13:29 -04:00

clip_vision_config_h.json

Add support for unCLIP SD2.x models.

2023-04-01 23:19:15 -04:00

clip_vision_config_vitl_336.json

support clip-vit-large-patch14-336 (#4042)

2024-07-17 13:12:50 -04:00

clip_vision_config_vitl.json

Add support for unCLIP SD2.x models.

2023-04-01 23:19:15 -04:00

clip_vision.py

support clip-vit-large-patch14-336 (#4042)

2024-07-17 13:12:50 -04:00

conds.py

Fix bug with cosxl edit model.

2024-06-10 16:00:03 -04:00

controlnet.py

Add SetUnionControlNetType to set the type of the union controlnet model.

2024-07-16 17:04:53 -04:00

diffusers_convert.py

Fix saving text encoder in fp8.

2024-04-02 11:46:34 -04:00

diffusers_load.py

Cleanup some unused imports.

2024-01-21 21:51:22 -05:00

gligen.py

Fix gligen lowvram mode.

2024-02-18 02:20:23 -05:00

latent_formats.py

Fast preview support for Flux.

2024-08-01 16:28:11 -04:00

lora.py

Hunyuan DiT lora support.

2024-07-25 22:42:54 -04:00

model_base.py

Automatically use fp8 for diffusion model weights if:

2024-08-03 13:45:19 -04:00

model_detection.py

Basic Flux Schnell and Flux Dev model implementation.

2024-08-01 09:49:29 -04:00

model_management.py

Automatically use fp8 for diffusion model weights if:

2024-08-03 13:45:19 -04:00

model_patcher.py

Add sampler_pre_cfg_function (#3979)

2024-07-09 16:20:49 -04:00

model_sampling.py

Basic Flux Schnell and Flux Dev model implementation.

2024-08-01 09:49:29 -04:00

ops.py

Lower T5 memory usage by a few hundred MB.

2024-07-31 00:52:34 -04:00

options.py

Only parse command line args when main.py is called.

2023-09-13 11:38:20 -04:00

sample.py

Reshape the empty latent image to the right amount of channels if needed.

2024-06-08 02:35:08 -04:00

sampler_helpers.py

Make ComfyUI split batches a higher priority than weight offload.

2024-08-01 16:39:59 -04:00

samplers.py

More aggressive batch splitting.

2024-08-03 11:53:30 -04:00

sd1_clip_config.json

Fix potential issue with non clip text embeddings.

2024-07-30 14:41:13 -04:00

sd1_clip.py

Lower CLIP memory usage by a bit.

2024-07-31 01:32:35 -04:00

sd.py

Automatically use fp8 for diffusion model weights if:

2024-08-03 13:45:19 -04:00

sdxl_clip.py

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

2024-07-25 10:52:09 -04:00

supported_models_base.py

Better per model memory usage estimations.

2024-08-02 18:09:24 -04:00

supported_models.py

Load T5 in fp8 if it's in fp8 in the Flux checkpoint.

2024-08-03 12:39:33 -04:00

types.py

Fix UnetParams annotation typo (#3589)

2024-05-27 19:30:35 -04:00

utils.py

Automatically use fp8 for diffusion model weights if:

2024-08-03 13:45:19 -04:00