search

comfyanonymous / ComfyUI

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.
https://www.comfy.org/
GNU General Public License v3.0
50.67k stars 5.32k forks source link

After updating to the latest version, flux nf4 becomes very slow #4538

Open martjay opened 3 weeks ago

martjay commented 3 weeks ago

Expected Behavior

A few days ago, I used the flux nf4 model to generate an image in just 1 minute.

Actual Behavior

Snipaste_2024-08-22_10-27-48 Snipaste_2024-08-22_10-28-07

Steps to Reproduce

flux NF4 V2.json My node has not changed, but the generation speed has become very slow.

Debug Logs

ComfyUI Launcher Diagnostic File

Date: 2024-08-22 10:50:48
Launcher Version: 2.8.9.363
Data File Version: 2024-08-20 06:52
ComfyUI Version: a60620dcea1302ef5c7f555e5e16f70b39c234ef (2024-08-22 04:38:26)
Working Directory: F:\AI\ComfyUI\ComfyUI-aki
------------------------
System Information: 
OS: Microsoft Windows NT 10.0.22631.0
CPU: 16 cores
Memory Size: 65536 MB
Page File Size: 10240 MB

NVIDIA Management Library:
  NVIDIA Driver Version: 560.70
  NVIDIA Management Library Version: 12.560.70

CUDA Driver:
  Version: 12060
  Devices: 
    00000000:01:00.0 0: NVIDIA GeForce RTX 2070 SUPER [75] 8 GB

NvApi:
  Version: 56070 r560_00

HIP Driver:
  Not Available

DirectML Driver: 
  Devices: 
    7812 0: NVIDIA GeForce RTX 2070 SUPER 7 GB
    7812 1: NVIDIA GeForce RTX 2070 SUPER 7 GB
    7812 2: NVIDIA GeForce RTX 2070 SUPER 7 GB
    7812 3: NVIDIA GeForce RTX 2070 SUPER 7 GB
    7812 4: NVIDIA GeForce RTX 2070 SUPER 7 GB

Intel Level Zero Driver:
  Not Available

------------------------
Environment Variables: 
COMPUTERNAME=BABA
BEATUNES_MEMORY=-Xmx640m
HOMEPATH=\Users\tianj
USERNAME=ColdPlay
EFC_7820=1
Path=C:\Users\tianj\AppData\Local\Programs\Python\Python311\Scripts\;C:\Users\tianj\AppData\Local\Programs\Python\Python311\;C:\Users\tianj\AppData\Local\Microsoft\WindowsApps;C:\ffmpeg\bin;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\lib\x64;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\include;C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1\extras\CUPTI\lib64;C:\Users\tianj\AppData\Local\Programs\Microsoft VS Code\bin;D:\BaoTa\jprq;C:\Users\tianj\AppData\Roaming\npm;C:\Users\tianj\AppData\Local\Programs\Ollama
HOMEDRIVE=C:
__PSLockDownPolicy=0
CommonProgramFiles=C:\Program Files\Common Files
ProgramFiles=C:\Program Files
ProgramData=C:\ProgramData
CommonProgramW6432=C:\Program Files\Common Files
DriverData=C:\Windows\System32\Drivers\DriverData
ALLUSERSPROFILE=C:\ProgramData
ProgramFiles(x86)=C:\Program Files (x86)
TEMP=C:\Users\tianj\AppData\Local\Temp
FPS_BROWSER_APP_PROFILE_STRING=Internet Explorer
KMP_WARNINGS=0
PROCESSOR_REVISION=a505
USERDOMAIN=BABA
CUDA_PATH_V12_1=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.1
APPDATA=C:\Users\tianj\AppData\Roaming
ChocolateyLastPathUpdate=133470854492414659
QT_DISABLE_SHADER_DISK_CACHE=1
QML_FORCE_DISK_CACHE=0
OneDrive=C:\Users\tianj\OneDrive
LOCALAPPDATA=C:\Users\tianj\AppData\Local
PAI_MODEL_DIR=D:\ProgramData\Topaz Labs LLC\Topaz Photo AI\models
ComSpec=C:\WINDOWS\system32\cmd.exe
RG_GPU_FRAMEWORK_ENGINE_RESOURCEDIR=C:\Program Files\Red Giant\RGFX
USERPROFILE=C:\Users\tianj
CommonProgramFiles(x86)=C:\Program Files (x86)\Common Files
QML_DISABLE_DISK_CACHE=1
NUMBER_OF_PROCESSORS=16
ProgramW6432=C:\Program Files
RG_UNIVERSE_FRAMEWORK_RESOURCEDIR=C:\Program Files\Red Giant\RGFX
ChocolateyInstall=C:\ProgramData\chocolatey
PATHEXT=.COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH;.MSC;.PY;.PYW
OS=Windows_NT
NVAFX_SDK_DIR=C:\Program Files\NVIDIA Corporation\NVIDIA Audio Effects SDK
ACE_STUDIO_PATH=D:\Program Files\ACE Studio
SESSIONNAME=Console
PROCESSOR_ARCHITECTURE=AMD64
INTEL_DEV_REDIST=C:\Program Files (x86)\Common Files\Intel\Shared Libraries\
LOGONSERVER=\\BABA
SystemDrive=C:
SystemRoot=C:\WINDOWS
windir=C:\WINDOWS
123pan=C:\Users\tianj\AppData\Local\123pan\123pan.exe
TMP=C:\Users\tianj\AppData\Local\Temp
PSModulePath=C:\Program Files\WindowsPowerShell\Modules;C:\WINDOWS\system32\WindowsPowerShell\v1.0\Modules
PROCESSOR_LEVEL=6
FPS_BROWSER_USER_PROFILE_STRING=Default
PROCESSOR_IDENTIFIER=Intel64 Family 6 Model 165 Stepping 5, GenuineIntel
PUBLIC=C:\Users\Public
USERDOMAIN_ROAMINGPROFILE=BABA
------------------------
Paths: 
Python: F:\AI\ComfyUI\ComfyUI-aki\python\python.exe
  - Python Paths:
    - F:\AI\ComfyUI\ComfyUI-aki\python\DLLs
    - F:\AI\ComfyUI\ComfyUI-aki\python\lib
    - F:\AI\ComfyUI\ComfyUI-aki\python
    - F:\AI\ComfyUI\ComfyUI-aki\python\lib\site-packages
    - F:\AI\ComfyUI\ComfyUI-aki\python\lib\site-packages\win32
    - F:\AI\ComfyUI\ComfyUI-aki\python\lib\site-packages\win32\lib
    - F:\AI\ComfyUI\ComfyUI-aki\python\lib\site-packages\Pythonwin
  - Python Packages:
    - absl_py 2.0.0 
    - accelerate 0.23.0 
    - addict 2.4.0 
    - aggdraw 1.3.18.post0 
    - aiohttp 3.8.5 
    - aiosignal 1.3.1 
    - albumentations 1.4.0 
    - annotated_types 0.7.0 
    - antlr4_python3_runtime 4.9.3 
    - anyio 3.7.1 
    - argostranslate 1.9.1 
    - arrow 1.3.0 
    - async_timeout 4.0.3 
    - attrs 23.1.0 
    - beautifulsoup4 4.12.2 
    - binaryornot 0.4.4 
    - bitsandbytes 0.43.3 
    - blind_watermark 0.4.4 
    - boltons 23.0.0 
    - cachetools 5.3.3 
    - certifi 2023.7.22 
    - cffi 1.16.0 
    - chardet 5.2.0 
    - charset_normalizer 3.2.0 
    - click 8.1.7 
    - clip_interrogator 0.6.0 
    - cmake 3.27.7 
    - colorama 0.4.6 
    - coloredlogs 15.0.1 
    - colorlog 6.8.0 
    - colour_science 0.4.4 
    - contourpy 1.1.1 
    - cookiecutter 2.6.0 
    - cryptography 42.0.5 
    - cssselect2 0.7.0 
    - ctranslate2 3.20.0 
    - cycler 0.12.1 
    - dashscope 1.20.4 
    - datasets 2.21.0 
    - decord 0.6.0 
    - deep_translator 1.11.4 
    - Deprecated 1.2.14 
    - diffusers 0.30.0 
    - dill 0.3.8 
    - distro 1.9.0 
    - easydict 1.13 
    - einops 0.6.1 
    - embreex 2.17.7.post4 
    - exceptiongroup 1.2.0 
    - fairscale 0.4.13 
    - fastapi 0.112.0 
    - filelock 3.12.4 
    - flatbuffers 23.5.26 
    - flet 0.23.2 
    - flet_core 0.23.2 
    - flet_runtime 0.23.2 
    - fonttools 4.43.1 
    - frozenlist 1.4.0 
    - fsspec 2023.9.1 
    - ftfy 6.1.1 
    - gdown 5.2.0 
    - gguf 0.9.1 
    - gitdb 4.0.11 
    - GitPython 3.1.40 
    - googleapis_common_protos 1.63.2 
    - google_ai_generativelanguage 0.6.6 
    - google_api_core 2.19.1 
    - google_api_python_client 2.140.0 
    - google_auth 2.33.0 
    - google_auth_httplib2 0.2.0 
    - google_generativeai 0.7.2 
    - gradio_client 1.3.0 
    - grpcio 1.65.4 
    - grpcio_status 1.48.2 
    - h11 0.14.0 
    - h2 4.1.0 
    - hpack 4.0.0 
    - httpcore 1.0.2 
    - httplib2 0.22.0 
    - httptools 0.6.1 
    - httpx 0.27.0 
    - huggingface_hub 0.23.4 
    - humanfriendly 10.0 
    - hyperframe 6.0.1 
    - idna 3.4 
    - imageio 2.35.0 
    - imageio_ffmpeg 0.5.1 
    - image_reward 1.5 
    - importlib_metadata 6.8.0 
    - Jinja2 3.1.2 
    - joblib 1.3.2 
    - jsonschema 4.20.0 
    - jsonschema_specifications 2023.7.1 
    - kiwisolver 1.4.5 
    - kornia 0.7.1 
    - lazy_loader 0.4 
    - llvmlite 0.41.1 
    - loguru 0.7.2 
    - lxml 4.9.3 
    - mapbox_earcut 1.0.1 
    - markdown_it_py 3.0.0 
    - MarkupSafe 2.1.3 
    - matplotlib 3.8.0 
    - matrix_client 0.4.0 
    - mdurl 0.1.1 
    - mediapipe 0.10.7 
    - mpmath 1.3.0 
    - multidict 6.0.4 
    - multiprocess 0.70.16 
    - networkx 3.1 
    - numba 0.58.1 
    - numexpr 2.8.8 
    - numpy 1.24.4 
    - oauthlib 3.2.2 
    - ollama 0.3.1 
    - omegaconf 2.3.0 
    - onnxruntime 1.17.1 
    - onnxruntime_gpu 1.16.3 
    - openai 1.2.4 
    - opencv_contrib_python 4.8.1.78 
    - opencv_python 4.9.0.80 
    - opencv_python_headless 4.7.0.72 
    - open_clip_torch 2.24.0 
    - packaging 23.1 
    - pandas 2.1.3 
    - piexif 1.1.3 
    - pilgram 1.2.1 
    - Pillow 9.5.0 
    - pip 23.0.1 
    - platformdirs 3.11.0 
    - pooch 1.8.0 
    - portalocker 2.8.2 
    - protobuf 3.20.3 
    - proto_plus 1.24.0 
    - psd_tools 1.9.34 
    - psutil 5.9.5 
    - pyarrow 17.0.0 
    - pyasn1 0.6.0 
    - pyasn1_modules 0.4.0 
    - pycparser 2.21 
    - pydantic 2.8.2 
    - pydantic_core 2.20.1 
    - PyGithub 2.3.0 
    - pygments 2.18.0 
    - PyJWT 2.8.0 
    - PyMatting 1.1.12 
    - PyNaCl 1.5.0 
    - pyparsing 3.1.1 
    - pypng 0.20220715.0 
    - pyreadline3 3.4.1 
    - PySocks 1.7.1 
    - python_dateutil 2.8.2 
    - python_dotenv 1.0.1 
    - python_slugify 8.0.4 
    - pytz 2023.3.post1 
    - pywavelets 1.5.0 
    - pywin32 306 
    - PyYAML 6.0.1 
    - pyzbar 0.1.9 
    - py_cpuinfo 9.0.0 
    - qrcode 7.4.2 
    - qudida 0.0.4 
    - referencing 0.30.2 
    - regex 2023.8.8 
    - rembg 2.0.52 
    - repath 0.9.0 
    - reportlab 4.0.6 
    - requests 2.32.3 
    - rich 13.7.1 
    - rpds_py 0.12.0 
    - rsa 4.9 
    - Rtree 1.1.0 
    - safetensors 0.4.2 
    - scikit_image 0.24.0 
    - scikit_learn 1.3.2 
    - scipy 1.12.0 
    - seaborn 0.13.0 
    - segment_anything 1.0 
    - sentencepiece 0.1.99 
    - setuptools 65.5.0 
    - shapely 2.0.2 
    - shellingham 1.5.4 
    - simpleeval 0.9.13 
    - six 1.16.0 
    - smmap 5.0.1 
    - sniffio 1.3.0 
    - sounddevice 0.4.6 
    - soundfile 0.12.1 
    - soupsieve 2.5 
    - spandrel 0.3.4 
    - stanza 1.1.1 
    - starlette 0.37.2 
    - svg.path 6.3 
    - sympy 1.12 
    - tabulate 0.9.0 
    - termcolor 2.3.0 
    - text_unidecode 1.3 
    - thop 0.1.1.post2209072238 
    - threadpoolctl 3.2.0 
    - tifffile 2023.9.26 
    - timm 0.6.13 
    - tinycss2 1.2.1 
    - tokenizers 0.19.1 
    - tomli 2.0.1 
    - torch 2.1.2+cu118 
    - torchaudio 2.1.2+cu118 
    - torchsde 0.2.5 
    - torchvision 0.16.2+cu118 
    - tqdm 4.66.5 
    - trampoline 0.1.2 
    - transformers 4.44.0 
    - transparent_background 1.3.1 
    - trimesh 4.0.9 
    - typer 0.12.3 
    - typer_config 1.4.0 
    - types_python_dateutil 2.9.0.20240316 
    - typing_extensions 4.8.0 
    - tzdata 2023.3 
    - ultralytics 8.0.223 
    - uritemplate 4.1.1 
    - urllib3 1.26.18 
    - uvicorn 0.30.6 
    - vhacdx 0.0.5 
    - watchdog 4.0.2 
    - watchfiles 0.23.0 
    - wcwidth 0.2.8 
    - webencodings 0.5.1 
    - websockets 12.0 
    - websocket_client 1.8.0 
    - win32_setctime 1.1.0 
    - wrapt 1.16.0 
    - xformers 0.0.23.post1+cu118 
    - xxhash 3.4.1 
    - yacs 0.1.8 
    - yapf 0.40.2 
    - yarl 1.9.2 
    - zhipuai 2.1.4.20230814 
    - zipp 3.17.0 
    - blend_modes 2.1.0 
    - cstr 0.1.0 
    - docopt 0.6.2 
    - ffmpy 0.3.0 
    - fvcore 0.1.5.post20221221 
    - googletrans_py 4.0.0 
    - img2texture 1.0.6 
    - iopath 0.1.10 
    - pycollada 0.8 
    - sacremoses 0.0.53 
    - svglib 1.5.1 
    - wget 3.2 
Git: F:\AI\ComfyUI\ComfyUI-aki\git\cmd\git.exe
Shell: C:\WINDOWS\system32\cmd.exe
Cache Path: F:\AI\ComfyUI\ComfyUI-aki\.cache
------------------------
Engine Validator: 
PyTorch: 2.1.2+cu118 C(11080)___
OnnxRuntime: CPU
------------------------
Port Info: 
Exclusion List:
  Non-administered:
    5357..1, 5426..1, 54235..1
  Administered:
    50000..60
In Use (v4):
  135 => 1840
  139 => 4
  445 => 4
  1337 => 9304
  3389 => 2104
  5040 => 8028
  5283 => 10808
  5354 => 7144
  5357 => 4
  5426 => 4
  7890 => 17120
  8090 => 15176
  8188 => 11744
  8680 => 1672
  9097 => 17120
  9197 => 6352
  11434 => 18888
  13013 => 1672
  13340 => 7216
  13344 => 7216
  33331 => 15424
  49668 => 1472
  49669 => 1364
  49670 => 2244
  49671 => 3712
  49672 => 3896
  49673 => 5212
  49709 => 1436
  49717 => 10228
  49741 => 11144
  49825 => 1740
  54235 => 4
  65001 => 7136
  65444 => 7872
In Use (v6):
  135 => 1840
  445 => 4
  3389 => 2104
  5357 => 4
  5426 => 4
  13337 => 9304
  49668 => 1472
  49669 => 1364
  49670 => 2244
  49671 => 3712
  49672 => 3896
  49673 => 5212
  49676 => 7092
  49709 => 1436
  54235 => 4
------------------------
Config: 
Audience Type: 专家
Engine: CUDA GPU 0: NVIDIA GeForce RTX 2070 SUPER (8 GB) [0]
VRAM Optimization: Auto [Auto]
Port: 8188 [8188]
XAttn Optimization: Xformers [Xformers]
Upcast Attention: True [True]
Precision: Auto [Auto]
Text Encoder Precision: Auto [Auto]
UNet Precision: Auto [Auto]
VAE Precision: Auto [Auto]
Preview Method: Auto [Auto]
Smart Memory: False [True]
Deterministic: False [False]
Listen: False [False]
Server Name:  []
HF Offline Mode: False [False]
Cuda Allocator Backend: Native [Native]
Prevent Sysmem Fallback: False
Extra Args: 
------------------------
Network Preferences: 
Proxy Address: http://127.0.0.1:7890
Proxy Git: True
Proxy Pip: True
Proxy Model Download: True
Proxy Env: True
Mirror Pypi: True
Mirror Git: True
Mirror ExtensionList: True
Mirror Huggingface: True
Github Acceleration: False
------------------------
Log: 
[START] Security scan
[DONE] Security scan
## ComfyUI-Manager: installing dependencies done.
** ComfyUI startup time: 
2024-08-22 10:29:07.111765
** Platform: 
Windows
** Python version: 
3.10.11 (tags/v3.10.11:7d4cc5a, Apr  5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
** Python executable: 
F:\AI\ComfyUI\ComfyUI-aki\python\python.exe
** ComfyUI Path: 
F:\AI\ComfyUI\ComfyUI-aki
** Log path: 
F:\AI\ComfyUI\ComfyUI-aki\comfyui.log

Prestartup times for custom nodes:
   0.0 seconds: 
F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\rgthree-comfy
   0.0 seconds: 
F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Marigold
   5.6 seconds: 
F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Manager

Total VRAM 8192 MB, total RAM 65425 MB
pytorch version: 2.1.2+cu118
xformers version: 0.0.23.post1+cu118
Set vram state to: NORMAL_VRAM
Disabling smart memory management
Device: cuda:0 NVIDIA GeForce RTX 2070 SUPER : native
Using xformers cross attention
[Prompt Server] web root: F:\AI\ComfyUI\ComfyUI-aki\web
Adding extra search path checkpoints D:/Downloads/2D/AI/aaaki-sd-webui/models/Stable-diffusion
Adding extra search path configs D:/Downloads/2D/AI/aaaki-sd-webui/models/Stable-diffusion
Adding extra search path vae D:/Downloads/2D/AI/aaaki-sd-webui/models/VAE
Adding extra search path loras D:/Downloads/2D/AI/aaaki-sd-webui/models/Lora
Adding extra search path loras D:/Downloads/2D/AI/aaaki-sd-webui/models/LyCORIS
Adding extra search path upscale_models D:/Downloads/2D/AI/aaaki-sd-webui/models/ESRGAN
Adding extra search path upscale_models D:/Downloads/2D/AI/aaaki-sd-webui/models/RealESRGAN
Adding extra search path upscale_models D:/Downloads/2D/AI/aaaki-sd-webui/models/SwinIR
Adding extra search path embeddings D:/Downloads/2D/AI/aaaki-sd-webui/embeddings
Adding extra search path hypernetworks D:/Downloads/2D/AI/aaaki-sd-webui/models/hypernetworks
Adding extra search path controlnet D:/Downloads/2D/AI/aaaki-sd-webui/models/ControlNet
[AnimateDiffEvo] - ERROR - No motion models found. Please download one and place in: ['F:\\AI\\ComfyUI\\ComfyUI-aki\\custom_nodes\\ComfyUI-AnimateDiff-Evolved\\models', 'F:\\AI\\ComfyUI\\ComfyUI-aki\\models\\animatediff_models']
Traceback (most recent call last):
  File "F:\AI\ComfyUI\ComfyUI-aki\nodes.py", line 1993, in load_custom_node
    module_spec.loader.exec_module(module)
  File "<frozen importlib._bootstrap_external>", line 883, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-GGUF\__init__.py", line 7, in <module>
    from .nodes import NODE_CLASS_MAPPINGS
  File "F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-GGUF\nodes.py", line 17, in <module>
    folder_paths.folder_names_and_paths["unet"][1].add(".gguf")
KeyError: 'unet'

Cannot import F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-GGUF module for custom nodes: 'unet'
### Loading: ComfyUI-Impact-Pack (V7.1.1)
[WARN] ComfyUI-Impact-Pack: custom_wildcards path not found: C:\aki\AAPACKING\ComfyUI-aki-v1\custom_nodes\ComfyUI-Impact-Pack\custom_wildcards. Using default path.
### Loading: ComfyUI-Impact-Pack (Subpack: V0.6)
[Impact Pack] Wildcards loading done.
### Loading: ComfyUI-Inspire-Pack (V0.86.1)
### Loading: ComfyUI-Manager (V2.50.1)
### ComfyUI Revision: 2594 [a60620dc] | Released on '2024-08-21'
### Loading: Workspace Manager (V1.0.0)

------------------------------------------

Comfyroll Studio v1.76 :  175 Nodes Loaded

------------------------------------------

** For changes, please see patch notes at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/blob/main/Patch_Notes.md

** For help, please see the wiki at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/wiki
------------------------------------------
[comfyui_controlnet_aux] | INFO -> Using ckpts path: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\comfyui_controlnet_aux\ckpts
[comfyui_controlnet_aux] | INFO -> Using symlinks: False
[comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider']
F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\comfyui_controlnet_aux\node_wrappers\dwpose.py:26: UserWarning: DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly
  warnings.warn("DWPose: Onnxruntime not found or doesn't come with acceleration providers, switch to OpenCV with CPU device. DWPose might run very slowly")
### [START] ComfyUI AlekPet Nodes ###
Node -> ArgosTranslateNode [Loading]

[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
[ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
Node -> ExtrasNode [Loading]
Node -> PainterNode [Loading]
Node -> PoseNode [Loading]

Node -> TranslateNode [Loading]

### [END] ComfyUI AlekPet Nodes ###
FizzleDorf Custom Nodes: Loaded
# 😺dzNodes: LayerStyle -> Cannot import name 'guidedFilter' from 'cv2.ximgproc'
A few nodes cannot works properly, while most nodes are not affected. Please REINSTALL package 'opencv-contrib-python'.
For detail refer to https://github.com/chflame163/ComfyUI_LayerStyle/issues/5
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\resource_dir.ini not found, default directory to be used. 
# 😺dzNodes: LayerStyle -> Find 1 LUTs in F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\lut
# 😺dzNodes: LayerStyle -> Find 1 Fonts in F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\font
Traceback (most recent call last):
  File "F:\AI\ComfyUI\ComfyUI-aki\nodes.py", line 1993, in load_custom_node
    module_spec.loader.exec_module(module)
  File "<frozen importlib._bootstrap_external>", line 883, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\efficiency-nodes-comfyui\__init__.py", line 9, in <module>
    from  .efficiency_nodes import NODE_CLASS_MAPPINGS
  File "F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 46, in <module>
    from .py import smZ_cfg_denoiser
  File "F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\efficiency-nodes-comfyui\py\smZ_cfg_denoiser.py", line 7, in <module>
    from comfy.samplers import KSampler, KSamplerX0Inpaint, wrap_model
ImportError: cannot import name 'wrap_model' from 'comfy.samplers' (F:\AI\ComfyUI\ComfyUI-aki\comfy\samplers.py)

Cannot import F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\efficiency-nodes-comfyui module for custom nodes: cannot import name 'wrap_model' from 'comfy.samplers' (F:\AI\ComfyUI\ComfyUI-aki\comfy\samplers.py)
Patching UNetModel.forward
UNetModel.forward has been successfully patched.
[Power Noise Suite]: 🦚🦚🦚 Cock-a-doodle-dooo! 🦚🦚🦚
[Power Noise Suite]: Tamed 11 wild nodes.

[rgthree] Loaded 34 exciting nodes.
[rgthree] NOTE: Will NOT use rgthree's optimized recursive execution as ComfyUI has changed.

WAS Node Suite: OpenCV Python FFMPEG support is enabled
WAS Node Suite Warning: `ffmpeg_bin_path` is not set in `F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\was-node-suite-comfyui\was_suite_config.json` config file. Will attempt to use system ffmpeg binaries if available.

WAS Node Suite: Finished. Loaded 218 nodes successfully.

    "Don't watch the clock; do what it does. Keep going." - Sam Levenson

Import times for custom nodes:
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\websocket_image_save.py
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\AIGODLIKE-ComfyUI-Translation
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\sdxl_prompt_styler
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-FluxExt-MZ
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\comfyui-workspace-manager
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\FreeU_Advanced
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ControlNet-LLLite-ComfyUI
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_TiledKSampler
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_Cutoff
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-WD14-Tagger
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_IPAdapter_plus
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_experiments
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-mape-Helpers
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\PowerNoiseSuite
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\images-grid-comfy-plugin
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-layerdiffusion
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Custom-Scripts
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Advanced-ControlNet
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_UltimateSDUpscale
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\Derfuu_ComfyUI_ModdedNodes
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\OneButtonPrompt
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\rgthree-comfy
   0.0 seconds (IMPORT FAILED): F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\efficiency-nodes-comfyui
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-AnimateDiff-Evolved
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Marigold
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_bitsandbytes_NF4
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\comfyui-ollama
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_Comfyroll_CustomNodes
   0.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\comfyui_controlnet_aux
   0.1 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_MiniCPM-V-2_6-int4
   0.1 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Inspire-Pack
   0.1 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-VideoHelperSuite
   0.1 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Long-CLIP
   0.3 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-layerdiffuse
   0.3 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_FizzNodes
   0.4 seconds (IMPORT FAILED): F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-GGUF
   0.4 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Manager
   0.7 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-LLMs
   0.7 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Impact-Pack
   2.2 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle
   5.9 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\was-node-suite-comfyui
   9.0 seconds: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_Custom_Nodes_AlekPet

Starting server

To see the GUI go to: http://127.0.0.1:8188
FETCH DATA from: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Manager\extension-node-map.json
 [DONE]
[Inspire Pack] IPAdapterPlus is not installed.
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 

got prompt
Failed to validate prompt for output 9:
* KSampler 3:
  - Required input is missing: negative
Output will be ignored
invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}
got prompt
Failed to validate prompt for output 9:
* KSampler 3:
  - Required input is missing: negative
Output will be ignored
invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}
FETCH DATA from: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI-Manager\extension-node-map.json [DONE]

# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
# 😺dzNodes: LayerStyle -> Warning: F:\AI\ComfyUI\ComfyUI-aki\custom_nodes\ComfyUI_LayerStyle\custom_size.ini not found, use default size. 
got prompt
Failed to validate prompt for output 9:
* CLIPTextEncode 12:
  - Required input is missing: clip
Output will be ignored
invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}
got prompt
model weight dtype torch.float16, manual cast: None
model_type FLUX
Using xformers attention in VAE
Using xformers attention in VAE
Requested to load FluxClipModel_
Loading 1 new model
loaded completely 0.0 4777.53759765625 True
Requested to load Flux
Loading 1 new model
loaded partially 5726.34351171875 6388.649485588074 0
Processing interrupted
Prompt executed in 305.66 seconds
Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLError(1, '[SSL: DECRYPTION_FAILED_OR_BAD_RECORD_MAC] decryption failed or bad record mac (_ssl.c:2578)')': /api/20/envelope/
Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'RemoteDisconnected('Remote end closed connection without response')': /api/20/envelope/
Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1007)')': /api/20/envelope/
Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1007)')': /api/20/envelope/
Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLEOFError(8, 'EOF occurred in violation of protocol (_ssl.c:1007)')': /api/20/envelope/
got prompt
Using xformers attention in VAE
Using xformers attention in VAE
Requested to load FluxClipModel_
Loading 1 new model
loaded completely 0.0 4777.53759765625 True
Requested to load Flux
Loading 1 new model
loaded partially 5645.496 6388.649485588074 0
Processing interrupted
------------------------
Fault Traceback: 
Not Available

Other

No response

DarkAlchy commented 3 weeks ago

I read somewhere where ComfyAnon has decided to make NF4 deprecated in favour of GGUF. I use NF4 myself, but haven't in the last few days (was nice and way faster than GGUF).

comfyanonymous commented 3 weeks ago

GGUF is a lot better, more flexible and will be faster than nf4 once someone ports the llama.cpp cuda kernels.

Your slowdown however is probably just because something else is using your vram.

martjay commented 3 weeks ago

GGUF is a lot better, more flexible and will be faster than nf4 once someone ports the llama.cpp cuda kernels.

Your slowdown however is probably just because something else is using your vram.

I don't know. I've seen that before starting to generate, there was almost 7GB VRAM. So I don't know if the change happened after updating Comfyui.

MrsHorrid commented 3 weeks ago

GGUF is a lot better, more flexible and will be faster than nf4 once someone ports the llama.cpp cuda kernels.

Your slowdown however is probably just because something else is using your vram.

any idea when the llama ccp kernles will be imported or is it just a sitting duck game?

Brostopher commented 3 weeks ago

GGUF is a lot better, more flexible and will be faster than nf4 once someone ports the llama.cpp cuda kernels. Your slowdown however is probably just because something else is using your vram.

I don't know. I've seen that before starting to generate, there was almost 7GB VRAM. So I don't know if the change happened after updating Comfyui.

Xformers was causing the issue for me. I only noticed a day or two after NF4 released. I thought it was a bug with NF4 but I had issues with across a lot of other workflows. Anytime I saw "Using xformers attention in VAE" it was like sitting at the DMV waiting for my number to be call.

I am only speaking from my experience and troubleshooting but hopefully one of the experts here confirm or deny my solution. Xformers removed support for PyTorch versions older than 2.2.0, so I updated PyTorch. Then installed the most current Xformers v0.0.27.post2. My gens are 4x faster than before.

Again, I have little understanding and using Comfy is my only experience with any type of coding but I hope this helps or points one of the gurus in the right direction for you.

Amit30swgoh commented 2 weeks ago

GGUF is a lot better, more flexible and will be faster than nf4 once someone ports the llama.cpp cuda kernels. Your slowdown however is probably just because something else is using your vram.

I don't know. I've seen that before starting to generate, there was almost 7GB VRAM. So I don't know if the change happened after updating Comfyui.

Xformers was causing the issue for me. I only noticed a day or two after NF4 released. I thought it was a bug with NF4 but I had issues with across a lot of other workflows. Anytime I saw "Using xformers attention in VAE" it was like sitting at the DMV waiting for my number to be call.

I am only speaking from my experience and troubleshooting but hopefully one of the experts here confirm or deny my solution. Xformers removed support for PyTorch versions older than 2.2.0, so I updated PyTorch. Then installed the most current Xformers v0.0.27.post2. My gens are 4x faster than before.

Again, I have little understanding and using Comfy is my only experience with any type of coding but I hope this helps or points one of the gurus in the right direction for you.

hey friend anyway i can contact you perhaps via Discord?

LUXQDee commented 2 weeks ago

It's not just NF4. Dev and Schnell have the same problem

martjay commented 2 weeks ago

GGUF is a lot better, more flexible and will be faster than nf4 once someone ports the llama.cpp cuda kernels. Your slowdown however is probably just because something else is using your vram.

I don't know. I've seen that before starting to generate, there was almost 7GB VRAM. So I don't know if the change happened after updating Comfyui.

Xformers was causing the issue for me. I only noticed a day or two after NF4 released. I thought it was a bug with NF4 but I had issues with across a lot of other workflows. Anytime I saw "Using xformers attention in VAE" it was like sitting at the DMV waiting for my number to be call.

I am only speaking from my experience and troubleshooting but hopefully one of the experts here confirm or deny my solution. Xformers removed support for PyTorch versions older than 2.2.0, so I updated PyTorch. Then installed the most current Xformers v0.0.27.post2. My gens are 4x faster than before.

Again, I have little understanding and using Comfy is my only experience with any type of coding but I hope this helps or points one of the gurus in the right direction for you.

Torch 2.3.1 (CUDA 12.1) + xFormers 0.0.27 not work for me