Sv. Lockal 0e0f1ed5bc

Allow bf16 computations on CPUs with BF16 support ...

Modern CPUs have native AVX512 BF16 instructions, which significantly improves matmul and conv2d operations.

With Bfloat16 instructions UNET steps are 40-50% faster on both AMD and Intel CPUs. There are minor visible changes with bf16, but no avalanche effects, so this feature is enabled by default with new `--use-cpu-bf16=auto` option. It can be disabled with `--use-cpu-bf16=no`.

Signed-off-by: Sv. Lockal <lockalsash@gmail.com>

2025-07-02 16:07:02 +00:00

..

cldm

Replace print with logging (#6138)

2024-12-20 16:24:55 -05:00

comfy_types

LoRA Trainer: LoRA training node in weight adapter scheme (#8446)

2025-06-13 19:25:59 -04:00

extra_samplers

Uni pc sampler now works with audio and video models.

2025-01-18 05:27:58 -05:00

image_encoders

Fix for dino lowvram. (#7748)

2025-04-23 04:12:52 -04:00

k_diffusion

Migrate ER-SDE from VE to VP algorithm and add its sampler node (#8744)

2025-07-01 02:38:52 -04:00

ldm

Don't add tiny bit of random noise when VAE encoding. (#8705)

2025-06-27 14:14:56 -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 contiguous issue with pytorch nightly. (#8729)

2025-06-29 06:38:40 -04:00

weight_adapter

LoRA Trainer: LoRA training node in weight adapter scheme (#8446)

2025-06-13 19:25:59 -04:00

checkpoint_pickle.py

Remove pytorch_lightning dependency.

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

cli_args.py

Allow bf16 computations on CPUs with BF16 support

2025-07-02 16:07:02 +00:00

clip_config_bigg.json

Fix potential issue with non clip text embeddings.

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

clip_model.py

Support llava clip vision model.

2025-03-06 00:24:43 -05:00

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_llava.json

Support llava clip vision model.

2025-03-06 00:24:43 -05:00

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_siglip_384.json

Support new flux model variants.

2024-11-21 08:38:23 -05:00

clip_vision_siglip_512.json

Support 512 siglip model.

2025-04-05 07:01:01 -04:00

clip_vision.py

throw away alpha channel in clip vision preprocessor (#7769)

2025-04-23 21:28:36 -04:00

conds.py

Make it easier to pass lists of tensors to models. (#8358)

2025-05-31 20:00:20 -04:00

controlnet.py

Alternate fix for #8435 (#8442)

2025-06-06 09:43:27 -04:00

diffusers_convert.py

Remove useless code.

2025-01-24 06:15:54 -05:00

diffusers_load.py

load_unet -> load_diffusion_model with a model_options argument.

2024-08-12 23:20:57 -04:00

float.py

Clamp output when rounding weight to prevent Nan.

2024-10-19 19:07:10 -04:00

gligen.py

Lint and fix undefined names (1/N) (#6028)

2024-12-12 18:55:26 -05:00

hooks.py

Hooks Part 2 - TransformerOptionsHook and AdditionalModelsHook (#6377)

2025-01-11 12:20:23 -05:00

latent_formats.py

Initial ACE-Step model implementation. (#7972)

2025-05-07 08:33:34 -04:00

lora_convert.py

Support the wan fun reward loras.

2025-04-07 05:01:47 -04:00

lora.py

Support HiDream SimpleTuner loras. (#8318)

2025-05-28 18:47:15 -04:00

model_base.py

Fix memory estimation bug with kontext. (#8709)

2025-06-27 17:21:12 -04:00

model_detection.py

Omnigen2 model implementation. (#8669)

2025-06-25 19:35:57 -04:00

model_management.py

Allow bf16 computations on CPUs with BF16 support

2025-07-02 16:07:02 +00:00

model_patcher.py

LoRA Trainer: LoRA training node in weight adapter scheme (#8446)

2025-06-13 19:25:59 -04:00

model_sampling.py

Basic initial support for cosmos predict2 text to image 2B and 14B models. (#8517)

2025-06-13 07:05:23 -04:00

ops.py

Fallback to regular op when fp8 op throws exception. (#8761)

2025-07-02 00:57:13 -04:00

options.py

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

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

patcher_extension.py

Add prepare_sampling wrapper allowing custom nodes to more accurately report noise_shape (#7500)

2025-04-09 09:43:35 -04:00

rmsnorm.py

Fix issue with old pytorch RMSNorm. (#8095)

2025-05-13 07:54:28 -04:00

sample.py

Auto reshape 2d to 3d latent for single image generation on video model.

2024-12-29 02:26:49 -05:00

sampler_helpers.py

Memory estimation code can now take into account conds. (#8307)

2025-05-27 15:09:05 -04:00

samplers.py

Reorder the schedulers so simple is the default one. (#8722)

2025-06-28 18:12:56 -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

Omnigen2 model implementation. (#8669)

2025-06-25 19:35:57 -04:00

sd.py

Omnigen2 model implementation. (#8669)

2025-06-25 19:35:57 -04:00

sdxl_clip.py

Add a T5TokenizerOptions node to set options for the T5 tokenizer. (#7803)

2025-04-25 19:36:00 -04:00

supported_models_base.py

Mixed precision diffusion models with scaled fp8.

2024-10-21 18:12:51 -04:00

supported_models.py

Disable omnigen2 fp16 on older pytorch versions. (#8672)

2025-06-26 03:39:09 -04:00

utils.py

Remove useless log. (#8166)

2025-05-17 06:27:34 -04:00