You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

comfyanonymous a094b45c93 Load clipvision model to GPU for faster performance.		2 years ago
..
cldm	Remove autocast from controlnet code.	2 years ago
extra_samplers	Refactor of sampler code to deal more easily with different model types.	2 years ago
k_diffusion	Fix referenced before assignment	2 years ago
ldm	Fallback to slice attention if xformers doesn't support the operation.	2 years ago
sd1_tokenizer	…
t2i_adapter	All resolutions now work with t2i adapter for SDXL.	2 years ago
taesd	…
checkpoint_pickle.py	…
cli_args.py	Enable bf16-vae by default on ampere and up.	2 years ago
clip_config_bigg.json	This is unused but it should be 1280.	2 years ago
clip_vision.py	Load clipvision model to GPU for faster performance.	2 years ago
clip_vision_config_g.json	Add support for clip g vision model to CLIPVisionLoader.	2 years ago
clip_vision_config_h.json	…
clip_vision_config_vitl.json	…
controlnet.py	Move ModelPatcher to model_patcher.py	2 years ago
diffusers_convert.py	Fix diffusers VAE loading.	2 years ago
diffusers_load.py	Support loading unet files in diffusers format.	2 years ago
gligen.py	Smarter memory management.	2 years ago
latent_formats.py	Support SDXL embedding format with 2 CLIP.	2 years ago
lora.py	Implement loras with norm keys.	2 years ago
model_base.py	Fix lowvram model merging.	2 years ago
model_detection.py	Support small diffusers controlnet so both types are now supported.	2 years ago
model_management.py	Enable bf16-vae by default on ampere and up.	2 years ago
model_patcher.py	Move ModelPatcher to model_patcher.py	2 years ago
ops.py	Initialize text encoder to target dtype.	2 years ago
sample.py	Try to free enough vram for control lora inference.	2 years ago
samplers.py	Support for Control Loras.	2 years ago
sd.py	Text encoder should initially load on the offload_device not the regular.	2 years ago
sd1_clip.py	Move text_projection to base clip model.	2 years ago
sd1_clip_config.json	…
sd2_clip.py	Initialize text encoder to target dtype.	2 years ago
sd2_clip_config.json	Move text_projection to base clip model.	2 years ago
sdxl_clip.py	Move text_projection to base clip model.	2 years ago
supported_models.py	Initialize the unet directly on the target device.	2 years ago
supported_models_base.py	Initialize the unet directly on the target device.	2 years ago
utils.py	Move controlnet code to comfy/controlnet.py	2 years ago