This workflow is giving me 416x736 resolution, 24FPS, 5 Second clips in 90-120 seconds (~75 seconds on gen, ~25 seconds on Decode, using the BF16 Model with multiple Loras. (Upscaling will add time) [If you can't use the Compiling or TeaCache you will get longer gen times.] It uses around 18GB of VRAM when generating.
I designed this specifically around a RTX 4090 - 24GB VRAM, 64GB RAM, running on Windows 10. I will not be able to provide recommendations for other hardware or operating systems.
I much prefer the outputs of the BF16 model of Hunyuan video, with the Triton Torch Compiling and TeaCache I get generations in 90-120 seconds (once the cache is warmed up) (Depending on your torch version you may have to make an edit to a PyTorch file, there is a guide in the notes in the workflow). While the model is loading it might lag the hell out of your PC but just be patient.
So far I found the best speeds and outputs at 416x736 resolution (or flipped), 121 frames (which at 24fps gives you 5 second clips) and steps between 20-30 (depending on your Loras)
Most Lora's I've tested work well between .5 and 1. You may need to lower the strength on one or more of the Loras when mixing them. If you get blurry/grainy videos try lowering.
Most of these discoveries have been trial and error, especially around the Triton Torch Compiling and TeaCache so I likely can't give much guidance outside of whats currently in the settings. Installing Triton on Windows was a nightmare - I used this guide: https://old.reddit.com/r/StableDiffusion/comments/1h7hunp/how_to_run_hunyuanvideo_on_a_single_24gb_vram_card/
None of this would be possible without kijai and the team behind the ComfyUI-HunyuanVideoWrapper so big props to them
Requires the following Custom Node Packages
Impact Pack - (For Wildcard Prompts) - https://github.com/ltdrdata/ComfyUI-Impact-Pack
KJNodes - (For Get/Set Nodes) - https://github.com/kijai/ComfyUI-KJNodes
HunyuanVideo Wrapper - (For Loading Hunyuan Models) - https://github.com/kijai/ComfyUI-HunyuanVideoWrapper (MAKE SURE YOU HAVE UPDATED SINCE 2025-01-18)
VideoHelperSuite (For creating video output) - https://github.com/Kosinkadink/ComfyUI-VideoHelperSuite
ComfyUI Frame Interpolation (For Frame Interpolation) - https://github.com/Fannovel16/ComfyUI-Frame-Interpolation
Requires the following Models
Hunyuan Video Safe Tensors - I like the BF16 best - https://civarchive.com/models/1018217?modelVersionId=1141746
Hunyuan Video VAE - https://civarchive.com/models/1018217?modelVersionId=1141674
Hunyuan Text Encoders - https://github.com/Tencent/HunyuanVideo/blob/main/ckpts/README.md#download-text-encoder
Upscale and Frame Interpolation Models
4x_foolhardy_Remacri - https://huggingface.co/FacehugmanIII/4x_foolhardy_Remacri
Film Net FP32 - https://civarchive.com/models/1167575?modelVersionId=1314385
If you find this Workflow help please consider donating buzz so I can further my Lora Trainings and Workflow development.
If you discover any improvements or issues please leave a comment on the post.
Description
This workflow is giving me 416x736 resolution, 24FPS, 5 Second clips in 90-120 seconds (~75 seconds on gen, ~25 seconds on Decode, using the BF16 Model with multiple Loras. (Upscaling will add time) [If you can't use the Compiling or TeaCache you will get longer gen times.] It uses around 18GB of VRAM when generating.
FAQ
Comments (52)
If you get
Error(s) in loading state_dict for AutoencoderKLCausal3D: Missing key(s) in state_dict: "encoder.down_blocks.0.resnets.0.norm1.weight", ....
You are probably using a VAE for the fp8 fastvideo model, which is the wrong one.
In the end though I could not get this to work on Linux with a 3090 and no OS (so 23.5GB usable VRAM). Quantization doesn't seem to want to work:
AssertionError: fp8e4nv data type is not supported on CUDA arch < 89
And every other option i tried (cudagraphs, flash_attn and so on) would not let me run because of OOM on VRAM.
Edit: ok, disable quantisation and enable blockswap, and it works. but the prompt execution sits at about 5 minutes.
Thanks for dropping that info here!
+1 on this comment. Ran into the same quantization issues. My poor 3090 :(
Same error. What is the fix? Disable quantisation and enable blockswap did not work.
@Minase460 So far no fix for the 30XX series. Seems to be a hardware compatibility issue.
Bummer. Seems like this requires Triton be installed and doesn't look like there's currently an easy way to do that in Pinokio/Comfy.
Yes, it's stated in the description, along with a guide on how to install Triton on Windows. It's not easy, but it's possible.
HyVideoTextEncode
list index out of range
error message constantly even though I have the required models & the 4090
Can you link an image of the HyVideoTextEncode node with how the settings are now? And does the console window/terminal show any longer messages.
My gut is that is either a misconfigured setting or something weird with the prompt length.
@TheAIDoctor :
I'll provide this to you tomorrow :) I'm finalizing a Lora training for Hunyuan video.
at the prompts level I am not affected by anything, I just loaded your workflow, indicated the correct elements: models, vae, etc.
and clicked: Launch
@AugusWrath Thanks!
@TheAIDoctor :
Good morning: as agreed together I will provide you with the error indicated by your workflow :)
# ComfyUI Error Report
## Error Details
- Node ID: 30
- Node Type: HyVideoTextEncode
- Exception Type: IndexError
- Exception Message: list index out of range
## Stack Trace
```
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
return_values = mapnode_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in mapnode_over_list
process_inputs(input_dict, i)
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\nodes.py", line 884, in process
prompt_embeds, negative_prompt_embeds, attention_mask, negative_attention_mask = encode_prompt(self,
^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\nodes.py", line 809, in encode_prompt
text_inputs = text_encoder.text2tokens(prompt,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\hyvideo\text_encoder\__init__.py", line 253, in text2tokens
text_tokens = self.processor(
^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\llava\processing_llava.py", line 145, in call
image_inputs = self.image_processor(images, **output_kwargs["images_kwargs"])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\image_processing_utils.py", line 41, in call
return self.preprocess(images, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\clip\image_processing_clip.py", line 286, in preprocess
images = make_list_of_images(images)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\image_utils.py", line 185, in make_list_of_images
if is_batched(images):
^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\image_utils.py", line 158, in is_batched
return is_valid_image(img[0])
~~~^^^
```
## System Information
- ComfyUI Version: 0.3.12
- Arguments: ComfyUI\main.py --windows-standalone-build
- OS: nt
- Python Version: 3.12.7 (tags/v3.12.7:0b05ead, Oct 1 2024, 03:06:41) [MSC v.1941 64 bit (AMD64)]
- Embedded Python: true
- PyTorch Version: 2.5.1+cu124
## Devices
- Name: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
- Type: cuda
- VRAM Total: 25756696576
- VRAM Free: 17421250210
- Torch VRAM Total: 6710886400
- Torch VRAM Free: 98774690
## Logs
```
2025-01-24T08:08:32.792049 - [Crystools [0;32mINFO[0m] Crystools version: 1.21.0
2025-01-24T08:08:32.820052 - [Crystools [0;32mINFO[0m] CPU: 13th Gen Intel(R) Core(TM) i7-13700KF - Arch: AMD64 - OS: Windows 11
2025-01-24T08:08:32.828049 - [Crystools [0;32mINFO[0m] Pynvml (Nvidia) initialized.
2025-01-24T08:08:32.828049 - [Crystools [0;32mINFO[0m] GPU/s:
2025-01-24T08:08:32.834050 - [Crystools [0;32mINFO[0m] 0) NVIDIA GeForce RTX 4090
2025-01-24T08:08:32.834050 - [Crystools [0;32mINFO[0m] NVIDIA Driver: 566.36
2025-01-24T08:08:35.014049 - [34m[ComfyUI-Easy-Use] server: [0mv1.2.7 [92mLoaded[0m2025-01-24T08:08:35.014049 -
2025-01-24T08:08:35.014049 - [34m[ComfyUI-Easy-Use] web root: [0mD:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Easy-Use\web_version/v2 [92mLoaded[0m2025-01-24T08:08:35.014049 -
2025-01-24T08:08:36.545049 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\timm\models\layers\__init__.py:48: FutureWarning: Importing from timm.models.layers is deprecated, please import via timm.layers
warnings.warn(f"Importing from {__name__} is deprecated, please import via timm.layers", FutureWarning)
2025-01-24T08:08:36.550052 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\timm\models\registry.py:4: FutureWarning: Importing from timm.models.registry is deprecated, please import via timm.models
warnings.warn(f"Importing from {__name__} is deprecated, please import via timm.models", FutureWarning)
2025-01-24T08:08:36.556051 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\timm\models\helpers.py:7: FutureWarning: Importing from timm.models.helpers is deprecated, please import via timm.models
warnings.warn(f"Importing from {__name__} is deprecated, please import via timm.models", FutureWarning)
2025-01-24T08:08:36.670052 - ### Loading: ComfyUI-Impact-Pack (V8.4.1)2025-01-24T08:08:36.670052 -
2025-01-24T08:08:36.790050 - ### Loading: ComfyUI-Impact-Subpack (V1.2.7)
2025-01-24T08:08:36.795050 - [Impact Pack] Wildcards loading done.2025-01-24T08:08:36.796050 -
2025-01-24T08:08:36.908051 - [Impact Subpack] ultralytics_bbox: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\ultralytics\bbox
2025-01-24T08:08:36.908051 - [Impact Subpack] ultralytics_segm: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\ultralytics\segm
2025-01-24T08:08:36.943050 - ----------Jake Upgrade Nodes Loaded----------2025-01-24T08:08:36.944050 -
2025-01-24T08:08:36.954050 - Total VRAM 24564 MB, total RAM 130918 MB
2025-01-24T08:08:36.954050 - pytorch version: 2.5.1+cu124
2025-01-24T08:08:36.954050 - Set vram state to: NORMAL_VRAM
2025-01-24T08:08:36.955049 - Device: cuda:0 NVIDIA GeForce RTX 4090 : cudaMallocAsync
2025-01-24T08:08:37.552050 - Please install pyav to use video processing functions.2025-01-24T08:08:37.552050 -
2025-01-24T08:08:38.067049 - [94mtheUpsiders Logic Nodes: [92mLoaded[0m2025-01-24T08:08:38.067049 -
2025-01-24T08:08:38.074050 - ### Loading: ComfyUI-Manager (V3.9.2)
2025-01-24T08:08:38.197051 - ### ComfyUI Version: v0.3.12-9-gd8a7a32 | Released on '2025-01-20'
2025-01-24T08:08:39.027050 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json
2025-01-24T08:08:39.041050 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json
2025-01-24T08:08:39.042051 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json
2025-01-24T08:08:39.071052 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json
2025-01-24T08:08:39.103050 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json
2025-01-24T08:08:40.489561 - --------------
2025-01-24T08:08:40.489561 - [91m ### Mixlab Nodes: [93mLoaded
2025-01-24T08:08:40.491561 - ChatGPT.available False
2025-01-24T08:08:40.492559 - edit_mask.available True
2025-01-24T08:08:40.507559 - ## clip_interrogator_model not found: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\clip_interrogator\Salesforce\blip-image-captioning-base, pls download from https://huggingface.co/Salesforce/blip-image-captioning-base2025-01-24T08:08:40.507559 -
2025-01-24T08:08:40.508558 - ClipInterrogator.available True
2025-01-24T08:08:40.509559 - ## text_generator_model not found: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\prompt_generator\text2image-prompt-generator, pls download from https://huggingface.co/succinctly/text2image-prompt-generator/tree/main2025-01-24T08:08:40.509559 -
2025-01-24T08:08:40.509559 - ## zh_en_model not found: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\prompt_generator\opus-mt-zh-en, pls download from https://huggingface.co/Helsinki-NLP/opus-mt-zh-en/tree/main2025-01-24T08:08:40.509559 -
2025-01-24T08:08:40.510559 - PromptGenerate.available True
2025-01-24T08:08:40.510559 - ChinesePrompt.available True
2025-01-24T08:08:40.510559 - RembgNode_.available True
2025-01-24T08:08:40.840559 - TripoSR.available
2025-01-24T08:08:40.841558 - MiniCPMNode.available
2025-01-24T08:08:41.913077 - Scenedetect.available
2025-01-24T08:08:41.991077 - FishSpeech.available
2025-01-24T08:08:42.036077 - SenseVoice.available
2025-01-24T08:08:42.096077 - Whisper.available False
2025-01-24T08:08:42.098077 - fal-client## OK2025-01-24T08:08:42.098077 -
2025-01-24T08:08:42.106079 - FalVideo.available
2025-01-24T08:08:42.107078 - [93m -------------- [0m
2025-01-24T08:08:42.345583 - Failed to auto update Quality of Life Suit 2025-01-24T08:08:42.345583 -
2025-01-24T08:08:42.346583 - [33mQualityOfLifeSuit_Omar92_DIR:[0m D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QualityOfLifeSuit_Omar922025-01-24T08:08:42.346583 -
2025-01-24T08:08:42.376584 - Traceback (most recent call last):
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2106, in load_custom_node
module_spec.loader.exec_module(module)
File "<frozen importlib._bootstrap_external>", line 995, in exec_module
File "<frozen importlib._bootstrap>", line 488, in callwith_frames_removed
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node-0.5.1\__init__.py", line 23, in <module>
from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node-0.5.1\nodes.py", line 15, in <module>
from insightface.app.common import Face
ModuleNotFoundError: No module named 'insightface'
2025-01-24T08:08:42.376584 - Cannot import D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node-0.5.1 module for custom nodes: No module named 'insightface'
2025-01-24T08:08:42.482583 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable2025-01-24T08:08:42.482583 -
2025-01-24T08:08:42.482583 - ############################################2025-01-24T08:08:42.482583 -
2025-01-24T08:08:42.482583 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\custom_nodes\ComfyUI-NAI-styler\CSV2025-01-24T08:08:42.482583 -
2025-01-24T08:08:42.483584 - ############################################2025-01-24T08:08:42.483584 -
2025-01-24T08:08:42.483584 - []2025-01-24T08:08:42.483584 -
2025-01-24T08:08:42.483584 - ############################################2025-01-24T08:08:42.483584 -
2025-01-24T08:08:42.771584 - ------------------------------------------2025-01-24T08:08:42.772583 -
2025-01-24T08:08:42.772583 - [34mComfyroll Studio v1.76 : [92m 175 Nodes Loaded[0m2025-01-24T08:08:42.772583 -
2025-01-24T08:08:42.772583 - ------------------------------------------2025-01-24T08:08:42.772583 -
2025-01-24T08:08:42.772583 - ** For changes, please see patch notes at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/blob/main/Patch_Notes.md2025-01-24T08:08:42.772583 -
2025-01-24T08:08:42.772583 - ** For help, please see the wiki at https://github.com/Suzie1/ComfyUI_Comfyroll_CustomNodes/wiki2025-01-24T08:08:42.772583 -
2025-01-24T08:08:42.772583 - ------------------------------------------2025-01-24T08:08:42.772583 -
2025-01-24T08:08:42.779584 - Initializing ControlAltAI Nodes2025-01-24T08:08:42.779584 -
2025-01-24T08:08:42.800583 - [36;20m[comfyui_controlnet_aux] | INFO -> Using ckpts path: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux\ckpts[0m
2025-01-24T08:08:42.801585 - [36;20m[comfyui_controlnet_aux] | INFO -> Using symlinks: False[0m
2025-01-24T08:08:42.801585 - [36;20m[comfyui_controlnet_aux] | INFO -> Using ort providers: ['CUDAExecutionProvider', 'DirectMLExecutionProvider', 'OpenVINOExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider', 'CoreMLExecutionProvider'][0m
2025-01-24T08:08:42.821584 - DWPose: Onnxruntime with acceleration providers detected2025-01-24T08:08:42.822583 -
2025-01-24T08:08:42.854583 - LOADED 6 FONTS2025-01-24T08:08:42.854583 -
2025-01-24T08:08:42.869584 - No OpenGL_accelerate module loaded: No module named 'OpenGL_accelerate'
2025-01-24T08:08:44.257094 -
███╗ ███╗ █████╗ ██████╗██╗ ██╗██╗███╗ ██╗███████╗
████╗ ████║██╔══██╗██╔════╝██║ ██║██║████╗ ██║██╔════╝
██╔████╔██║███████║██║ ███████║██║██╔██╗ ██║█████╗
██║╚██╔╝██║██╔══██║██║ ██╔══██║██║██║╚██╗██║██╔══╝
██║ ╚═╝ ██║██║ ██║╚██████╗██║ ██║██║██║ ╚████║███████╗
╚═╝ ╚═╝╚═╝ ╚═╝ ╚═════╝╚═╝ ╚═╝╚═╝╚═╝ ╚═══╝╚══════╝
██████╗ ███████╗██╗ ██╗ ██╗███████╗██╗ ██████╗ ███╗ ██╗███████╗
██╔══██╗██╔════╝██║ ██║ ██║██╔════╝██║██╔═══██╗████╗ ██║██╔════╝
██║ ██║█████╗ ██║ ██║ ██║███████╗██║██║ ██║██╔██╗ ██║███████╗
██║ ██║██╔══╝ ██║ ██║ ██║╚════██║██║██║ ██║██║╚██╗██║╚════██║
██████╔╝███████╗███████╗╚██████╔╝███████║██║╚██████╔╝██║ ╚████║███████║
╚═════╝ ╚══════╝╚══════╝ ╚═════╝ ╚══════╝╚═╝ ╚═════╝ ╚═╝ ╚═══╝╚══════╝
2025-01-24T08:08:44.257094 -
2025-01-24T08:08:44.325604 - # 😺dzNodes: LayerStyle -> [1;33mCannot 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 [4mhttps://github.com/chflame163/ComfyUI_LayerStyle/issues/5[0m[m2025-01-24T08:08:44.325604 -
2025-01-24T08:08:44.460603 - [92m[tinyterraNodes] [32mLoaded[0m2025-01-24T08:08:44.460603 -
2025-01-24T08:08:44.530604 - Adding2025-01-24T08:08:44.530604 - 2025-01-24T08:08:44.530604 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes2025-01-24T08:08:44.530604 - 2025-01-24T08:08:44.531604 - to sys.path2025-01-24T08:08:44.531604 -
2025-01-24T08:08:44.538604 - Traceback (most recent call last):
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2106, in load_custom_node
module_spec.loader.exec_module(module)
File "<frozen importlib._bootstrap_external>", line 995, in exec_module
File "<frozen importlib._bootstrap>", line 488, in callwith_frames_removed
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\init .py", line 25, in <module>
from .modules.nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS
ImportError: attempted relative import with no known parent package
2025-01-24T08:08:44.538604 - Cannot import D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\init .py module for custom nodes: attempted relative import with no known parent package
2025-01-24T08:08:45.051603 - [34mWAS Node Suite: [0mOpenCV Python FFMPEG support is enabled[0m2025-01-24T08:08:45.051603 -
2025-01-24T08:08:45.051603 - [34mWAS Node Suite [93mWarning: [0m`ffmpeg_bin_path` is not set in D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\pr-was-node-suite-comfyui-47064894\was_suite_config.json config file. Will attempt to use system ffmpeg binaries if available.[0m2025-01-24T08:08:45.051603 -
2025-01-24T08:08:45.251604 - FETCH ComfyRegistry Data: 5/302025-01-24T08:08:45.251604 -
2025-01-24T08:08:45.529605 - [34mWAS Node Suite: [0mFinished.[0m [32mLoaded[0m [0m218[0m [32mnodes successfully.[0m2025-01-24T08:08:45.529605 -
2025-01-24T08:08:45.529605 -
[3m[93m"The harder I work, the luckier I get."[0m[3m - Samuel Goldwyn[0m
2025-01-24T08:08:45.529605 -
2025-01-24T08:08:45.576605 -
2025-01-24T08:08:45.576605 - [92m[rgthree-comfy] Loaded 42 fantastic nodes. 🎉[00m2025-01-24T08:08:45.576605 -
2025-01-24T08:08:45.576605 -
2025-01-24T08:08:46.145112 - Searge-SDXL v4.3.1 in D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL2025-01-24T08:08:46.145112 -
2025-01-24T08:08:46.178112 -
Import times for custom nodes:
2025-01-24T08:08:46.178112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py
2025-01-24T08:08:46.178112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-ImageMotionGuider
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-unload-model
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoImagesGuider
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-mxToolkit
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-logic
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoMultiLora
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfy-cliption
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Logic-1.2.0
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\sdxl_prompt_styler
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\Comfyui_TTP_Toolset-1.0.0
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\cg-use-everywhere
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-copilot
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-YOLO
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_TiledKSampler
2025-01-24T08:08:46.179112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_ttp_toolset
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Universal-Styler
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\stability-ComfyUI-nodes
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\cg-image-picker
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\masquerade-nodes-comfyui
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyLiterals
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_ipadapter_plus
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlaltai_nodes
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\Comfy-WaveSpeed
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\mikey_nodes
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_essentials
2025-01-24T08:08:46.180112 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-various
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyMath
2025-01-24T08:08:46.181113 - 0.0 seconds (IMPORT FAILED): D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\init .py
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfy-image-saver
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Chibi-Nodes
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\derfuu_comfyui_moddednodes
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Custom-Scripts
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-advanced-controlnet
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Frame-Interpolation
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-florence2
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-QualityOfLifeSuit_Omar92
2025-01-24T08:08:46.181113 - 0.0 seconds (IMPORT FAILED): D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node-0.5.1
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Comfyroll_CustomNodes
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-kjnodes
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-jakeupgrade
2025-01-24T08:08:46.181113 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_smznodes
2025-01-24T08:08:46.182114 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\rgthree-comfy
2025-01-24T08:08:46.182114 - 0.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoStyler
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\aegisflow_utility_nodes
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_tinyterraNodes
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Isulion
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_LayerStyle
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-tensorops
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-dream-project
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-impact-subpack
2025-01-24T08:08:46.182114 - 0.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-impact-pack
2025-01-24T08:08:46.182114 - 0.2 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-ollama
2025-01-24T08:08:46.182114 - 0.3 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-VideoHelperSuite
2025-01-24T08:08:46.182114 - 0.5 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LLaVA-OneVision
2025-01-24T08:08:46.182114 - 0.6 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-kynode
2025-01-24T08:08:46.183113 - 0.6 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfy-mtb
2025-01-24T08:08:46.183113 - 0.6 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\SeargeSDXL
2025-01-24T08:08:46.183113 - 0.6 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-detail-daemon
2025-01-24T08:08:46.183113 - 0.7 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Easy-Use
2025-01-24T08:08:46.183113 - 0.7 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-dynamicprompts
2025-01-24T08:08:46.183113 - 0.9 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager
2025-01-24T08:08:46.183113 - 1.0 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\pr-was-node-suite-comfyui-47064894
2025-01-24T08:08:46.183113 - 1.2 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Crystools
2025-01-24T08:08:46.183113 - 1.4 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_fill-nodes
2025-01-24T08:08:46.183113 - 1.5 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GIMM-VFI
2025-01-24T08:08:46.183113 - 3.1 seconds: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes
2025-01-24T08:08:46.183113 -
2025-01-24T08:08:46.198112 - Starting server
2025-01-24T08:08:46.198112 - To see the GUI go to: http://127.0.0.1:8188
2025-01-24T08:08:49.566584 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/photoswipe-lightbox.esm.min.js2025-01-24T08:08:49.566584 -
2025-01-24T08:08:49.595580 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/photoswipe.min.css2025-01-24T08:08:49.595580 -
2025-01-24T08:08:49.619575 - FETCH DATA from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\extension-node-map.json2025-01-24T08:08:49.619575 - 2025-01-24T08:08:49.632575 - [DONE]2025-01-24T08:08:49.632575 -
2025-01-24T08:08:49.638582 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/pickr.min.js2025-01-24T08:08:49.639578 -
2025-01-24T08:08:49.702577 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/classic.min.css2025-01-24T08:08:49.702577 -
2025-01-24T08:08:50.035325 - [33mQualityOfLifeSuit_Omar92:[0m:NSP ready2025-01-24T08:08:50.035325 -
2025-01-24T08:08:50.035325 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.035325 -
2025-01-24T08:08:50.035325 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.036328 -
2025-01-24T08:08:50.036328 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.036328 -
2025-01-24T08:08:50.036328 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.036328 -
2025-01-24T08:08:50.036328 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.036328 -
2025-01-24T08:08:50.036328 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.036328 -
2025-01-24T08:08:50.465407 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/model-viewer.min.js2025-01-24T08:08:50.465407 -
2025-01-24T08:08:50.513414 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/juxtapose.css2025-01-24T08:08:50.513414 -
2025-01-24T08:08:50.530412 - D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-mixlab-nodes\webApp\lib/juxtapose.min.js2025-01-24T08:08:50.530412 -
2025-01-24T08:08:50.790427 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.790427 -
2025-01-24T08:08:50.790427 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.791428 -
2025-01-24T08:08:50.791428 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.791428 -
2025-01-24T08:08:50.791428 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.791428 -
2025-01-24T08:08:50.791428 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.791428 -
2025-01-24T08:08:50.791428 - Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh".
Your current root directory is: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI
2025-01-24T08:08:50.791428 -
2025-01-24T08:08:51.690020 - FETCH ComfyRegistry Data: 10/302025-01-24T08:08:51.690020 -
2025-01-24T08:08:57.576362 - FETCH ComfyRegistry Data: 15/302025-01-24T08:08:57.579361 -
2025-01-24T08:09:04.096899 - FETCH ComfyRegistry Data: 20/302025-01-24T08:09:04.096899 -
2025-01-24T08:09:10.455931 - FETCH ComfyRegistry Data: 25/302025-01-24T08:09:10.456931 -
2025-01-24T08:09:16.680813 - FETCH ComfyRegistry Data: 30/302025-01-24T08:09:16.680813 -
2025-01-24T08:09:17.181613 - FETCH ComfyRegistry Data [DONE]2025-01-24T08:09:17.181613 -
2025-01-24T08:09:17.209618 - [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes
2025-01-24T08:09:17.238624 - nightly_channel: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/remote
2025-01-24T08:09:17.238624 - FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json2025-01-24T08:09:17.238624 - 2025-01-24T08:09:17.303322 - [DONE]2025-01-24T08:09:17.303322 -
2025-01-24T08:09:32.992837 - got prompt
2025-01-24T08:09:32.998846 - Failed to validate prompt for output 34:
2025-01-24T08:09:32.998846 - * HyVideoLoraSelect 46:
2025-01-24T08:09:32.998846 - - Value not in list: lora: 'hunyaun\amat_petitie\edge_of_reality_ep30_25200stp_lora_theaidoctor.safetensors' not in (list of length 209)
2025-01-24T08:09:32.998846 - Output will be ignored
2025-01-24T08:09:32.998846 - invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}
2025-01-24T08:09:43.579411 - got prompt
2025-01-24T08:09:45.813013 - Loading text encoder model (clipL) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\clip\clip-vit-large-patch14
2025-01-24T08:09:47.150633 - Text encoder to dtype: torch.float16
2025-01-24T08:09:47.239641 - Loading tokenizer (clipL) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\clip\clip-vit-large-patch14
2025-01-24T08:09:47.544112 - Loading text encoder model (llm) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\LLM\llava-llama-3-8b-text-encoder-tokenizer
2025-01-24T08:09:47.823372 -
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]2025-01-24T08:09:47.824377 -
Loading checkpoint shards: 0%| | 0/4 [00:00<?, ?it/s]2025-01-24T08:09:47.824377 -
2025-01-24T08:09:47.824377 - !!! Exception during processing !!! No such file or directory: "D:\\ComfyUI_windows_portable_nvidia\\ComfyUI_windows_portable\\ComfyUI\\models\\LLM\\llava-llama-3-8b-text-encoder-tokenizer\\model-00001-of-00004.safetensors"
2025-01-24T08:09:47.894378 - Traceback (most recent call last):
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
return_values = mapnode_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in mapnode_over_list
process_inputs(input_dict, i)
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\nodes.py", line 684, in loadmodel
text_encoder = TextEncoder(
^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\hyvideo\text_encoder\__init__.py", line 167, in init
self.model, self.model_path = load_text_encoder(
^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\hyvideo\text_encoder\__init__.py", line 39, in load_text_encoder
text_encoder = AutoModel.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\auto\auto_factory.py", line 564, in from_pretrained
return model_class.from_pretrained(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4224, in from_pretrained
) = cls._load_pretrained_model(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 4770, in loadpretrained_model
state_dict = load_state_dict(
^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\modeling_utils.py", line 504, in load_state_dict
with safe_open(checkpoint_file, framework="pt") as f:
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
FileNotFoundError: No such file or directory: "D:\\ComfyUI_windows_portable_nvidia\\ComfyUI_windows_portable\\ComfyUI\\models\\LLM\\llava-llama-3-8b-text-encoder-tokenizer\\model-00001-of-00004.safetensors"
2025-01-24T08:09:47.895376 - Prompt executed in 4.31 seconds
2025-01-24T08:10:16.491469 - got prompt
2025-01-24T08:10:16.504469 - Loading text encoder model (clipL) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\clip\clip-vit-large-patch14
2025-01-24T08:10:16.693741 - Text encoder to dtype: torch.float16
2025-01-24T08:10:16.808742 - Loading tokenizer (clipL) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\clip\clip-vit-large-patch14
2025-01-24T08:10:16.880742 - Using a slow image processor as use_fast is unset and a slow processor was saved with this model. use_fast=True will be the default behavior in v4.48, even if the model was saved with a slow processor. This will result in minor differences in outputs. You'll still be able to use a slow processor with use_fast=False.
2025-01-24T08:10:16.890742 - You are using the default legacy behaviour of the <class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>. This is expected, and simply means that the legacy (previous) behavior will be used so nothing changes for you. If you want to use the new behaviour, set legacy=False. This should only be set if you understand what it means, and thoroughly read the reason why this was added as explained in https://github.com/huggingface/transformers/pull/24565 - if you loaded a llama tokenizer from a GGUF file you can ignore this message.
2025-01-24T08:10:17.627905 - Loading text encoder model (vlm) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\LLM\llava-llama-3-8b-v1_1-transformers
2025-01-24T08:11:05.428659 -
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████| 4/4 [00:47<00:00, 10.62s/it]2025-01-24T08:11:05.428659 -
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████| 4/4 [00:47<00:00, 11.90s/it]2025-01-24T08:11:05.428659 -
2025-01-24T08:11:05.657658 - Text encoder to dtype: torch.float16
2025-01-24T08:11:05.657658 - Loading tokenizer (vlm) from: D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\models\LLM\llava-llama-3-8b-v1_1-transformers
2025-01-24T08:11:06.016658 - !!! Exception during processing !!! list index out of range
2025-01-24T08:11:06.033660 - Traceback (most recent call last):
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data
return_values = mapnode_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in mapnode_over_list
process_inputs(input_dict, i)
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs
results.append(getattr(obj, func)(**inputs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\nodes.py", line 884, in process
prompt_embeds, negative_prompt_embeds, attention_mask, negative_attention_mask = encode_prompt(self,
^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\nodes.py", line 809, in encode_prompt
text_inputs = text_encoder.text2tokens(prompt,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\hyvideo\text_encoder\__init__.py", line 253, in text2tokens
text_tokens = self.processor(
^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\llava\processing_llava.py", line 145, in call
image_inputs = self.image_processor(images, **output_kwargs["images_kwargs"])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\image_processing_utils.py", line 41, in call
return self.preprocess(images, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\models\clip\image_processing_clip.py", line 286, in preprocess
images = make_list_of_images(images)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\image_utils.py", line 185, in make_list_of_images
if is_batched(images):
^^^^^^^^^^^^^^^^^^
File "D:\ComfyUI_windows_portable_nvidia\ComfyUI_windows_portable\python_embeded\Lib\site-packages\transformers\image_utils.py", line 158, in is_batched
return is_valid_image(img[0])
~~~^^^
IndexError: list index out of range
2025-01-24T08:11:06.035660 - Prompt executed in 49.54 seconds
```
## Attached Workflow
Please make sure that workflow does not contain any sensitive information such as API keys or passwords.
```
Workflow too large. Please manually upload the workflow from local file system.
```
## Additional Context
(Please add any additional context or steps to reproduce the error here)
@AugusWrath There's quite a few errors indicated in there but it appears to stem from the text2tokens function within the ComfyUI-HunyuanVideoWrapper node
The error log indicates missing model files such as:
No such file or directory: "D:\\ComfyUI_windows_portable_nvidia\\ComfyUI_windows_portable\\ComfyUI\\models\\LLM\\llava-llama-3-8b-text-encoder-tokenizer\\model-00001-of-00004.safetensors"
So I would make sure that model is present in your file system (or just try reselecting the text encoder in the llm model selector)
@TheAIDoctor it works I will test and I will keep you informed :) thank you
@TheAIDoctor :
Good morning I come to give you some news, so I re-downloaded the models, re-selected the right textvideo encoder and still the same thing, I have 0 worries on other workflows whether made by myself or by other users, there is only on yours or the models and the textencoder indicates this error to me.
I do not understand why this workflow refuses there or others work without worries, I did not dare and know how to modify the workflow.
and is it possible to make a ( ) variant or alternative via another textencoder?
thank you for your return
@AugusWrath did u fix it
@miraishounen thx
DownloadAndLoadHyVideoTextEncoder
Failed to import transformers.models.timm_wrapper.configuration_timm_wrapper because of the following error (look up to see its traceback): cannot import name 'ImageNetInfo' from 'timm.data'
Seems like one of you packages didn't install a pip package correctly. Specifically "ImageNetInfo" you could try manually installing that packages. Or it may be easier to just try to update all of the required custom node modules.
How long's the torch compile time? I found it too time consuming and taxing on my system RAM for it to be usable.
Takes about 2ish minutes I think on the first warmup for the model + lora. Then its reused until you restart comfy or change the model / lora config.
Might work on 30** cards, but I wasn't successful.
Got Triton installed, managed to build sageattention as well; but ran into some wall that I don't feel like breaking down right now.
Seems like Triton isn't fully supporting 30** cards at this time.
Great effort by the creator though. appreciate the help!
FYI This only works on 40** cards. I tried it on my 3090, but due to triton not being able to run quantisation on 30** cards, it's hella slow.
That doesn't seem right to me, I was under the impression triton had no issues with the 30 series as I've seen numerous threads where people have had success with it.
https://github.com/triton-inference-server/server/issues/2299
https://www.reddit.com/r/StableDiffusion/comments/1g45n6n/triton_3_wheels_published_for_windows_and_working/ls4d4pb/
https://civitai.com/articles/8384/guide-to-installing-8-bit-attention-sage-for-comfy-ui
@TheAIDoctor Triton does work. Maybe i misunderstood; but the quantization in the model loader doesn't work. Not entirely sure how that impacts the flow; but it was stuck for a good 5 minutes before I gave up.. I'll try and be more patient and report back.
@veldhuisnick771 You could try lowering the Dynamo Cache Size Limit to 32 or 16. I'm not sure what effect that will have on the speed gain or quality but it may compile faster.
Additionally you can experiment with:
fullgraph = true: Optimize the entire graph to reduce memory fragmentation.
dynamic = true: For models with varying input sizes or batch dimensions, this helps reduce wasted memory.
Finally you could last try:
The cudagraphs backend which an be faster in some cases, but it comes with constraints:
- The input sizes must remain consistent because CUDA Graphs cannot handle dynamic shapes.
- Memory allocation patterns must be fixed for it to work efficiently.
Which I'm not sure of the backend scripting for that Node if it has appropriate handling the limitations.
@veldhuisnick771 You can also try different quantization on the model loader. Historically the default selected in the workflow works best for me but I've seen other workflows using different quant methods.
@TheAIDoctor I've been messing about with it; but haven't gotten it to work yet. Took me a solid 4 hours to get sageattention compiled. lmao.
Right now i'm runningit, it's been some 20 minutes and still compiling. Python is doing some stuff in the background, hovering around 10% cpu usage. between 13-16gb ram, and 23.7gb of gpu ram.
Visually there isn't anything happening. No messages in the cli either.
I've tried several quantization modes, but non seemed to work. though I couldnt trie the torchao; it errored I didnt have it installed.
https://github.com/woct0rdho/triton-windows#fp8-is-not-supported-on-rtx-30xx-and-older-gpus
This is the comment from Triton Windows about it.
@Itsnameless What quant method do you normally use on other Hunyuan workflows?
@TheAIDoctor Honestly? Beats me. I've been using the Hun tea advanced workflow flawlessly with great results. I think it uses the FP8 one, but doesn't do it through triton.
In any case, I appreciate your help so far, but I think it's a lost cause for now, until triton somehow manages to get 3090 in the game.
https://i.ibb.co/gyS82Gf/cmd-RVDbymh7u-H.png
Lmao
@Itsnameless Bummer but understandable! I'll drop a comment here if I ever figure out how it can work for a 3090 in the future.
Thanks for testing these out! I also have a 3090, so will be curious if this ever works. Those times are insane!
was able to install triton with https://github.com/woct0rdho/triton-windows/releases/tag/v3.2.0-windows.post9
But final error on 3090 is quite obvious with fp8 :
RuntimeError: torch._scaled_mm is only supported on CUDA devices with compute capability >= 9.0 or 8.9, or ROCm MI300+
Everything works, but I'm seeing this:
The config attributes {'use_flow_sigmas': True, 'prediction_type': 'flow_prediction'} were passed to FlowMatchDiscreteScheduler, but are not expected and will be ignored. Please verify your scheduler_config.json configuration file.
Yeah it always shows that for me as well. I've just ignored it without issue. I think it's a mismatch or error in the custom nodes.
Okay TY
For anyone else on Windows who encounters this -
torch._dynamo.exc.BackendCompilerFailed: backend='inductor' raised: FileNotFoundError: [Errno 2] No such file or directory
You need to "Enable Win32 long paths" via group policy editor - Local Computer Policy > Computer Configuration > Administrative Templates > System > Filesystem.
The issue being that the temp inductor folder path was being truncated, thus not found after being created. Took me hours...
Works beautifully now.
Good find. I've had that setting on for years so I forget about it.
Is it normal to have to reload the hunyuan model on every generation? 64gb ram 4090.
Nope, that's not normal. It should only reload the model if you change LORA [strength] or some other operation that changes how the model needs to be loaded.
There are certain node settings such as "force offload" that could push it off. It can also offload if you run out of VRAM. For example if you were at 23 out of 24gb of VRAM use at the video sampler step and then the decode or an upscale step after pushed you over the 24gb it could unload the model to make sure the next steps can run.
Is it possible to select the Sampler in this workflow. One LORA recommends 'gradient_estimation' versus 'Euler' but I don't see where that can be selected.
SOLVED:
Just trying this tonight and keep getting this error, tried searching for an answer but still struggling. Any info would be appreciated,
Thank you :)
DownloadAndLoadHyVideoTextEncoder
No package metadata was found for bitsandbytes
It does show it as installed of I try to install it, but the report shows this:
ComfyUI Error Report
## Error Details
- Node ID: 16
- Node Type: DownloadAndLoadHyVideoTextEncoder
- Exception Type: importlib.metadata.PackageNotFoundError
- Exception Message: No package metadata was found for bitsandbytes
## Stack Trace
```
File "G:\- Stability Matrix -\Data\Packages\ComfyUI\execution.py", line 327, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "G:\- Stability Matrix -\Data\Packages\ComfyUI\execution.py", line 202, in get_output_data
return_values = mapnode_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "G:\- Stability Matrix -\Data\Packages\ComfyUI\execution.py", line 174, in mapnode_over_list
process_inputs(input_dict, i)
File "G:\- Stability Matrix -\Data\Packages\ComfyUI\execution.py", line 163, in process_inputs
results.append(getattr(obj, func)(**inputs))
File "G:\- Stability Matrix -\Data\Packages\ComfyUI\custom_nodes\comfyui-hunyuanvideowrapper\nodes.py", line 671, in loadmodel
quantization_config = BitsAndBytesConfig(
File "G:\- Stability Matrix -\Data\Packages\ComfyUI\venv\lib\site-packages\transformers\utils\quantization_config.py", line 419, in init
ok, so I removed it and re-installed it and it's working now, so maybe an update, not sure - At least if anyone has this issue, this might help : )
And now I have TORCH errors :(
I don't think this will run on a Windows PC.
@skpManiac I run it on windows. But you need WSL and triton installed to properly run everything like TeaCache and the torch compile
@TheAIDoctor Thanks for the heads up :)
HyVideoTextEncode
list index out of range
Worked great for me, Arch linux 4090. Buzz sent, keep up the good work!
Thanks so much for the tip! I'll have a new version up in the next few days. Been designing some custom nodes to make things a little cleaner.
Working on a 3090 after changing quantization = fp8_e5m2 . Triton functions were compiling fine with your original settings until hitting
Triton compilation failed: triton_poi_fused__to_copy_1
Then I found the workaround advice here:
https://github.com/kijai/ComfyUI-HunyuanVideoWrapper/issues/116#issuecomment-2671321903
So I have a suggestion for your next iteration, there is a node called "unload model" if you place this after the conditioning and put the clip model into its model input, this will unload the model after the prompt is encoded. This will clear up a chunk of ram, then unload the HYV model after the intermediary, and unload ALL at the end of the upscale, I also place a load image (any image) -> model unload all -> preview image as the very first model. it may be redundant but if i stop it short , or switch workflows without model unloading, it ensures it's always clear. And I went from 3 second videos 10m @ 7 steps to 10second videos 1:08m @ 7 steps , my 3090 went to gpu heaven so I'm working with a 2080 and the 12gb doesn't like doing much, until now.. it's quite quick, works great with flux too, unloading t5 is a hefty speed boost since you clear up that vram,