Post "1-Click INSTALL V3 ULTIMATE FLUX & VIDEO GENERATION ComfyUI WebUI!" by "Aitrepreneur" from patreon

Smog Lebottier

Hello, I managed to install ComfyUI and the workflow on a RunPod instance, however I get an error each time I try to generate a picture: "Loop Detected 110,399,394, " for FP8 Gen and "Loop Detected 503,512,510, " for GGUF. Can anyone help me troubleshot this please?

pietje puk

It looks like an amazing workflow for comfy. Unfortunately it doesn't work on a 12 GB 3080 Ti. It keeps complaining Sampler Custom Advanced Allocation on device. I just just the txt2img workflow and the lowest memory setting (Q4)

Aitrepreneur

You installed it as admin hence why it's installed inside the system32 folder and why you got errors, install it normally

Michael

I'm on Windows 11, fresh install. I ran V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL-UPDATED-M.bat I have a 3060 12Gb, so I ran it with option 1 selected, then told it to install the Schnell model, and don't install the FP8 model. It took a while so I fell asleep. I woke up the next day and it was installed, so I closed the cmd terminal window, shut the computer down and had breakfast. Then I came back and ran it. When I hit Run I got permission denied when it tried to save the output image to C:\Windows\System32\ComfyUI_windows_portable\ComfyUI\output I also can't save the workflow. It gives permission denied. I watched some videos, and tried following directions more closely. I clicked the manager, then update all, and got permissions denied again. It looks like ComfyUI can't read or write to anything in the ComfyUI folders. I tried fiddling with the folder perms in Windows 11, to give apps read/write permission, but that didn't help. Also, the discord link doesn't work, so I can't ask for help there. This seems like the only place I can ask for help. Any advice?

tramidus mondge

hello, when i execute flux-lora-fluxgym-install-v2.bat, i have an error : Checking for Rust toolchain.... Rust not found, installing into a temporary directory [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. [notice] A new release of pip is available: 24.2 -> 25.1.1 [notice] To update, run: python.exe -m pip install --upgrade pip error: metadata-generation-failed × Encountered error while generating package metadata. ╰─> See above for output. note: This is an issue with the package mentioned above, not pip. hint: See above for details.

lepomeduusa

Read the latest comments on this post, there are some compability issues with the latest ComfyUI and Aitrpreneurs install files. There are several non-pro people who've tried a ton of different things myself included, some of them work some don't so we're waiting for the actual pro to fix this but Aitpreneur has some personal things that won't allow him to be present 24/7 so we have to be a bit patient. I'm sure he'll answer as soon as he has the mental and physical capacity. :)

Minh Đặng

Hi everyone, newbie here. I restart my pod. Then go back to port 3000 but Comfy does not open. How do I properly open/restart comfy ui on runpod? Can someone help me out. Thanks.

rubz

for days im trying to get it working again, somehow it appears comfyui changes nodes vallues autmaticly and randomly uppon saving them and recaling them, like batch number going insane high so it breaks with out of memory, also the upscaler Florence 2 nodes break randomly. not sure if its comfy or nodes that need updating,, please help Aintrepeneur

Edgar

Hello I launched V3 all in one on runpod, followed everything by instructions, even went through the troubleshooting guide, BUT at the end, when I installed all nodes, I have this ERROR "self2.canvas.ds.offset is not iterable" when trying to use the workflow

lepomeduusa

Update from my side (working with Runpod): I got some things (mind you only a few essentials) working by: - First updated Comfy to the latest batch - Then ran the .sh inside Jupyter - Refreshed Comfy and installed / updated rest of the nodes The issue is that it's not 100 % fix. Sometimes after the installation I get "self2.canvas.ds.offset is not iterable" which as far as I know is something to do with the custom nodes breaking the flow. And every time after these steps HunyuanVideoWrapper, PuLID_Flux_II and PuLID_Flux_Enhanced are still missing and no matter how many times you try to fix them, uninstall and reinstall they won't work.

Nathan Blake

Where do I find the batch size? mind sharing a screenshot?

Tiago Fernandes

Everything was working flawless... And now it is completely broken. I'll investigate further and I'll update the comment here asap.

lepomeduusa

Hey man! Just wanted to say huuuuge thanks for all the work you do. You make everything easy to understand and my own knowledge of working with Comfy has improved a ton. I'm here to ask of the same issues some other people have had that I didn't previously have and it seems to be because of a Comfy update. "torch.OutOfMemoryError:" > got this working somehow by reducing batch quantity. The main issue I have right now is, that when I do a complete clean new Comfy in Runpod and update it, it seems to break your workflow both in Jupyter and inside Comfy > some missing nodes that don't seem to work even though I've tried everything. Would love your input on this, is there any easy way to fix this that I don't get..? :) Some reddit and Github convos say that the new update separates the back end and front end and that causes some issues.

Dougomite

I'm just back to once again say thanks for doing all this and putting it all together in a simple to use installer. "torch.OutOfMemoryError:" I wanted to call out an issue I ran into today, when I loaded up the All in one workflow. The Batch Size of everything was set to 512 which gave me out of memory errors if I tried to use any of the generators. I had to drop all the Batch Sizes down to a handful to get things running again. Just wanted to mention it in case someone else has the same issue.

rubz

The previous comfy broke some stuff but i sort off got it working in latest v0.3.30 now. I would appreciate a v4 version which is consolidate a little bit, removing some out of date stuff and perhaps split the ALL IN ONE in 2 seperate workflows , ALL IN ONE image generation and ALL IN ONE video generation, because the video models are released pretty quickly every week one ore more new models are being released .. cheers!

Martin Kurth

For me as a newbie easy to install on runpod. My problem is that I get "import failed" for HunyuanVideoWrapper, PuLID_Flux_II and PuLID_Flux_Enhanced, whatever I do. Uninstalling, Reinstalling, "Try fix" all doesn't help. The log says something like: File "/usr/local/lib/python3.11/dist-packages/diffusers/utils/import_utils.py", line 810, in __getattr__ module = self._get_module(self._class_to_module[name]) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/dist-packages/diffusers/utils/import_utils.py", line 822, in _get_module raise RuntimeError( RuntimeError: Failed to import diffusers.models.autoencoders.autoencoder_kl because of the following error (look up to see its traceback): Cannot set attribute 'src' directly. Use '_unsafe_update_src()' and manually clear `.hash` of all callersinstead.

shamiir goldsmith

Why doesn't RunPodt accept the drag-and-drop update? This is the last step; it's so frustrating

Blue

I'm having the same issue. Missing nodes that wont install with the Install missing nodes option.

Vincent Nestler

PULIDREDUX - this node does not show when I try to download missing nodes. Thoughts?

Iggix

The updated runpod installer is failing: When loading the graph, the following node types were not found: LTXAttnOverride (In group node 'workflow/LTXPARAMTEXTOVIDGGUF') workflow/LTXPARAMTEXTOVIDGGUFRemove from workflow ModifyLTXModel (In group node 'workflow/LTXPARAMTEXTOVIDGGUF') LTXPerturbedAttention (In group node 'workflow/LTXPARAMTEXTOVIDGGUF') LTXVConditioning (In group node 'workflow/LTXPARAMTEXTOVIDGGUF') workflow/LTXPARAMTXTFPRemove from workflow workflow/LTXIMAGETOVRemove from workflow LTXVImgToVideo (In group node 'workflow/LTXIMAGETOV') LTXVScheduler (In group node 'workflow/LTXIMAGETOV') workflowsampler and seed upscaler workflowExtra nodes upscaler workflowFlorence workflowFlux Controlnet workflowImage gen sampler workflowOther Nodes workflowOther Nodes 2 workflowSampler 3 workflowNodes 3 workflowParam workflowsampler img workflowNegative Prompt workflowXLabs Sampler workflowpositive prompt guidance workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullidpara workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowpostiveguid workflowDiff workflowoutpaintpar workflowREDUX1 workflowREDUX2 workflowREDUX3 workflowPULIDREDUX workflowREDUXPARAM workflowREDUXOTHERPARAM workflownewcontrolnetparam workflowshaker workflowshakercont workflowSHAKKERPARAM workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM EmptyLTXVLatentVideo workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhynupar EmptyHunyuanLatentVideo workflowhuynyav2vparm HyVideoModelLoader workflowreduxnormparam workflowprompinredux Nodes that have failed to load will show as red on the graph.

JpaLVVmI6c8xM00

I'm getting the same error. Is there a fix?

Aitrepreneur

send me a dm but I can't replicate any issue on my side, everything works even with a brand new install

J

Hey, there must've been updates or something because your updated .bat no longer downloads all the models in one go as well as the v3_1 workflow .json continues to spit out the same missing node types as all the other users experiencing the error. (starting from florence all the way to the huanyuan nodes) Some missing downloads were mitigated by having to manually auth/ssh into hugging-face. I was using your workflow all the time and now it's unfortunately no longer functioning out the gate. :( If you need dev assistance I'd be happy to brainstorm your bottlenecks

Rluu

Why I can't use anime pony model on fp8?

Rluu

is anyone here have this error? "mat1 and mat2 shapes cannot be multiplied (1x2304 and 2816x1280)"

Rafik Biff

I have the same problem, how to fix it please ?

Andrew Elliott

With the auto installer, is there a way to change the default installation directory. I tried using it, but it defaulted to my C drive and ran out of space

benoit corvol

Thank's for the tips !

Chad Lumley

Love the work you're doing. Just a quick optimization: for the runpod shell script, I'd recommend updating the script to download the dependencies first then calling a separate python script that downloads the models, that way you can utilize hf_transfer to speed up the model downloads. I got a fast pod yesterday and was getting 1GB/s transfers.

Daniele Petta

è possibile installare il programma su un altro disco esempio (E:) ? perchè il mio disco (C:) è full

Hometrees

officially solved. I don't even know what i did but i solved it! lmao what a ride gotta love comfy

Hometrees

nevermind still running into an issue gosh darn it all Prompt outputs failed validation UnetLoaderGGUF: - Value not in list: unet_name: 'fast-hunyuan-video-t2v-720p-Q8_0.gguf' not in []

Hometrees

SOLVED super user error like monkey brain level user error, if you come across this error it just means your .sh file ISNT IN YOUR COMFYUI FOLDER ON RUNPOD go back and move it to that folder. lmao so sorry for commenting about such a silly user error

Hometrees

I tried on the GGUF side but get this error on runpod Prompt outputs failed validation DualCLIPLoader: - Value not in list: clip_name1: 'clip_l.safetensors' not in [] - Value not in list: clip_name2: 'llava_llama3_fp8_scaled.safetensors' not in [] UnetLoaderGGUF: - Required input is missing: unet_name VAELoader: - Value not in list: vae_name: 'hunyuan_video_vae_bf16.safetensors' not in ['sdxl_vae.safetensors'] any way i can manually download these models?

Carl Nimbus

I am also seeing the same even after running Linux auto install script do to a fresh install. Even after restarting comfy and dragging the v3 or the v3.1 workflows in, same error.

Hometrees

ah man i used the gguf side and still got this error Prompt outputs failed validation DualCLIPLoader: - Value not in list: clip_name1: 'clip_l.safetensors' not in [] - Value not in list: clip_name2: 'llava_llama3_fp8_scaled.safetensors' not in [] UnetLoaderGGUF: - Value not in list: unet_name: 'fast-hunyuan-video-t2v-720p-Q8_0.gguf' not in [] VAELoader: - Value not in list: vae_name: 'hunyuan_video_vae_bf16.safetensors' not in ['sdxl_vae.safetensors'] It seems the models are not installing ------- as for the other error let me grab it for you -------- root@8c257067672e:/workspace# chmod +x V3-ULTIMATE-FLUX_AUTO_INSTALL-RUNPOD-UPDATED.sh ./V3-ULTIMATE-FLUX_AUTO_INSTALL-RUNPOD-UPDATED.sh Collecting insightface Downloading insightface-0.7.3.tar.gz (439 kB) Installing build dependencies ... done Getting requirements to build wheel ... done Preparing metadata (pyproject.toml) ... done Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from insightface) (1.26.3) Collecting onnx (from insightface) Downloading onnx-1.17.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (16 kB) Requirement already satisfied: tqdm in /usr/local/lib/python3.11/dist-packages (from insightface) (4.66.4) Requirement already satisfied: requests in /usr/local/lib/python3.11/dist-packages (from insightface) (2.32.3) Collecting matplotlib (from insightface) Downloading matplotlib-3.10.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (11 kB) Requirement already satisfied: Pillow in /usr/local/lib/python3.11/dist-packages (from insightface) (10.2.0) Collecting scipy (from insightface) Downloading scipy-1.15.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (61 kB) Collecting scikit-learn (from insightface) Downloading scikit_learn-1.6.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (18 kB) Collecting scikit-image (from insightface) Downloading scikit_image-0.25.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (14 kB) Collecting easydict (from insightface) Downloading easydict-1.13-py3-none-any.whl.metadata (4.2 kB) Collecting cython (from insightface) Using cached Cython-3.0.12-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (3.3 kB) Collecting albumentations (from insightface) Downloading albumentations-2.0.5-py3-none-any.whl.metadata (41 kB) Collecting prettytable (from insightface) Downloading prettytable-3.15.1-py3-none-any.whl.metadata (33 kB) Requirement already satisfied: PyYAML in /usr/local/lib/python3.11/dist-packages (from albumentations->insightface) (6.0.1) Collecting pydantic>=2.9.2 (from albumentations->insightface) Downloading pydantic-2.10.6-py3-none-any.whl.metadata (30 kB) Collecting albucore==0.0.23 (from albumentations->insightface) Downloading albucore-0.0.23-py3-none-any.whl.metadata (5.3 kB) Collecting opencv-python-headless>=4.9.0.80 (from albumentations->insightface) Downloading opencv_python_headless-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (20 kB) Collecting stringzilla>=3.10.4 (from albucore==0.0.23->albumentations->insightface) Downloading stringzilla-3.12.2-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl.metadata (80 kB) Collecting simsimd>=5.9.2 (from albucore==0.0.23->albumentations->insightface) Downloading simsimd-6.2.1-cp311-cp311-manylinux_2_28_x86_64.whl.metadata (66 kB) Collecting contourpy>=1.0.1 (from matplotlib->insightface) Downloading contourpy-1.3.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (5.4 kB) Collecting cycler>=0.10 (from matplotlib->insightface) Downloading cycler-0.12.1-py3-none-any.whl.metadata (3.8 kB) Collecting fonttools>=4.22.0 (from matplotlib->insightface) Downloading fonttools-4.56.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (101 kB) Collecting kiwisolver>=1.3.1 (from matplotlib->insightface) Downloading kiwisolver-1.4.8-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.2 kB) Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.11/dist-packages (from matplotlib->insightface) (24.1) Requirement already satisfied: pyparsing>=2.3.1 in /usr/lib/python3/dist-packages (from matplotlib->insightface) (2.4.7) Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.11/dist-packages (from matplotlib->insightface) (2.9.0.post0) Collecting protobuf>=3.20.2 (from onnx->insightface) Downloading protobuf-6.30.0-cp39-abi3-manylinux2014_x86_64.whl.metadata (593 bytes) Requirement already satisfied: wcwidth in /usr/local/lib/python3.11/dist-packages (from prettytable->insightface) (0.2.13) Requirement already satisfied: charset-normalizer<4,>=2 in /usr/local/lib/python3.11/dist-packages (from requests->insightface) (3.3.2) Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.11/dist-packages (from requests->insightface) (3.7) Requirement already satisfied: urllib3<3,>=1.21.1 in /usr/local/lib/python3.11/dist-packages (from requests->insightface) (2.2.2) Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.11/dist-packages (from requests->insightface) (2024.7.4) Requirement already satisfied: networkx>=3.0 in /usr/local/lib/python3.11/dist-packages (from scikit-image->insightface) (3.2.1) Collecting imageio!=2.35.0,>=2.33 (from scikit-image->insightface) Downloading imageio-2.37.0-py3-none-any.whl.metadata (5.2 kB) Collecting tifffile>=2022.8.12 (from scikit-image->insightface) Downloading tifffile-2025.2.18-py3-none-any.whl.metadata (31 kB) Collecting lazy-loader>=0.4 (from scikit-image->insightface) Downloading lazy_loader-0.4-py3-none-any.whl.metadata (7.6 kB) Collecting joblib>=1.2.0 (from scikit-learn->insightface) Downloading joblib-1.4.2-py3-none-any.whl.metadata (5.4 kB) Collecting threadpoolctl>=3.1.0 (from scikit-learn->insightface) Downloading threadpoolctl-3.5.0-py3-none-any.whl.metadata (13 kB) Collecting annotated-types>=0.6.0 (from pydantic>=2.9.2->albumentations->insightface) Downloading annotated_types-0.7.0-py3-none-any.whl.metadata (15 kB) Collecting pydantic-core==2.27.2 (from pydantic>=2.9.2->albumentations->insightface) Downloading pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.6 kB) Collecting typing-extensions>=4.12.2 (from pydantic>=2.9.2->albumentations->insightface) Downloading typing_extensions-4.12.2-py3-none-any.whl.metadata (3.0 kB) Requirement already satisfied: six>=1.5 in /usr/lib/python3/dist-packages (from python-dateutil>=2.7->matplotlib->insightface) (1.16.0) Downloading albumentations-2.0.5-py3-none-any.whl (290 kB) Downloading albucore-0.0.23-py3-none-any.whl (14 kB) Downloading scipy-1.15.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (37.6 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 37.6/37.6 MB 185.6 MB/s eta 0:00:00 Using cached Cython-3.0.12-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.6 MB) Downloading easydict-1.13-py3-none-any.whl (6.8 kB) Downloading matplotlib-3.10.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (8.6 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.6/8.6 MB 203.4 MB/s eta 0:00:00 Downloading onnx-1.17.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (16.0 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 16.0/16.0 MB 213.1 MB/s eta 0:00:00 Downloading prettytable-3.15.1-py3-none-any.whl (33 kB) Downloading scikit_image-0.25.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (14.8 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 14.8/14.8 MB 116.8 MB/s eta 0:00:00 Downloading scikit_learn-1.6.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.5 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.5/13.5 MB 223.9 MB/s eta 0:00:00 Downloading contourpy-1.3.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (326 kB) Downloading cycler-0.12.1-py3-none-any.whl (8.3 kB) Downloading fonttools-4.56.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (4.9 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 4.9/4.9 MB 252.1 MB/s eta 0:00:00 Downloading imageio-2.37.0-py3-none-any.whl (315 kB) Downloading joblib-1.4.2-py3-none-any.whl (301 kB) Downloading kiwisolver-1.4.8-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.4 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.4/1.4 MB 221.6 MB/s eta 0:00:00 Downloading lazy_loader-0.4-py3-none-any.whl (12 kB) Downloading opencv_python_headless-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (50.0 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 50.0/50.0 MB 243.2 MB/s eta 0:00:00 Downloading protobuf-6.30.0-cp39-abi3-manylinux2014_x86_64.whl (316 kB) Downloading pydantic-2.10.6-py3-none-any.whl (431 kB) Downloading pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.0 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.0/2.0 MB 183.6 MB/s eta 0:00:00 Downloading threadpoolctl-3.5.0-py3-none-any.whl (18 kB) Downloading tifffile-2025.2.18-py3-none-any.whl (226 kB) Downloading annotated_types-0.7.0-py3-none-any.whl (13 kB) Downloading simsimd-6.2.1-cp311-cp311-manylinux_2_28_x86_64.whl (632 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 632.7/632.7 kB 139.3 MB/s eta 0:00:00 Downloading stringzilla-3.12.2-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl (307 kB) Downloading typing_extensions-4.12.2-py3-none-any.whl (37 kB) Building wheels for collected packages: insightface Building wheel for insightface (pyproject.toml) ... done Created wheel for insightface: filename=insightface-0.7.3-cp311-cp311-linux_x86_64.whl size=1064925 sha256=c9e3665eb486f8f5a72511ca35ad30a7b582fd2d8817969ee2c9bfdba4bc55c8 Stored in directory: /root/.cache/pip/wheels/27/d8/22/f52d858d16cd06e7b2e6aad34a1777dcfaf000be833bbf8146 Successfully built insightface Installing collected packages: stringzilla, simsimd, easydict, typing-extensions, tifffile, threadpoolctl, scipy, protobuf, prettytable, opencv-python-headless, lazy-loader, kiwisolver, joblib, imageio, fonttools, cython, cycler, contourpy, annotated-types, scikit-learn, scikit-image, pydantic-core, onnx, matplotlib, albucore, pydantic, albumentations, insightface Attempting uninstall: typing-extensions Found existing installation: typing_extensions 4.9.0 Uninstalling typing_extensions-4.9.0: Successfully uninstalled typing_extensions-4.9.0 Successfully installed albucore-0.0.23 albumentations-2.0.5 annotated-types-0.7.0 contourpy-1.3.1 cycler-0.12.1 cython-3.0.12 easydict-1.13 fonttools-4.56.0 imageio-2.37.0 insightface-0.7.3 joblib-1.4.2 kiwisolver-1.4.8 lazy-loader-0.4 matplotlib-3.10.1 onnx-1.17.0 opencv-python-headless-4.11.0.86 prettytable-3.15.1 protobuf-6.30.0 pydantic-2.10.6 pydantic-core-2.27.2 scikit-image-0.25.2 scikit-learn-1.6.1 scipy-1.15.2 simsimd-6.2.1 stringzilla-3.12.2 threadpoolctl-3.5.0 tifffile-2025.2.18 typing-extensions-4.12.2 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Collecting facexlib Downloading facexlib-0.3.0-py3-none-any.whl.metadata (4.6 kB) Collecting filterpy (from facexlib) Downloading filterpy-1.4.5.zip (177 kB) Installing build dependencies ... done Getting requirements to build wheel ... done Preparing metadata (pyproject.toml) ... done Collecting numba (from facexlib) Downloading numba-0.61.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.8 kB) Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from facexlib) (1.26.3) Collecting opencv-python (from facexlib) Downloading opencv_python-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (20 kB) Requirement already satisfied: Pillow in /usr/local/lib/python3.11/dist-packages (from facexlib) (10.2.0) Requirement already satisfied: scipy in /usr/local/lib/python3.11/dist-packages (from facexlib) (1.15.2) Requirement already satisfied: torch in /usr/local/lib/python3.11/dist-packages (from facexlib) (2.4.0+cu121) Requirement already satisfied: torchvision in /usr/local/lib/python3.11/dist-packages (from facexlib) (0.19.0+cu121) Requirement already satisfied: tqdm in /usr/local/lib/python3.11/dist-packages (from facexlib) (4.66.4) Requirement already satisfied: matplotlib in /usr/local/lib/python3.11/dist-packages (from filterpy->facexlib) (3.10.1) Collecting llvmlite<0.45,>=0.44.0dev0 (from numba->facexlib) Downloading llvmlite-0.44.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.8 kB) Requirement already satisfied: filelock in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (3.13.1) Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (4.12.2) Requirement already satisfied: sympy in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (1.12) Requirement already satisfied: networkx in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (3.2.1) Requirement already satisfied: jinja2 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (3.1.3) Requirement already satisfied: fsspec in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (2024.2.0) Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (12.1.105) Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (12.1.105) Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (12.1.105) Requirement already satisfied: nvidia-cudnn-cu12==9.1.0.70 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (9.1.0.70) Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (12.1.3.1) Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (11.0.2.54) Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (10.3.2.106) Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (11.4.5.107) Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (12.1.0.106) Requirement already satisfied: nvidia-nccl-cu12==2.20.5 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (2.20.5) Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (12.1.105) Requirement already satisfied: triton==3.0.0 in /usr/local/lib/python3.11/dist-packages (from torch->facexlib) (3.0.0) Requirement already satisfied: nvidia-nvjitlink-cu12 in /usr/local/lib/python3.11/dist-packages (from nvidia-cusolver-cu12==11.4.5.107->torch->facexlib) (12.1.105) Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.11/dist-packages (from jinja2->torch->facexlib) (2.1.5) Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy->facexlib) (1.3.1) Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy->facexlib) (0.12.1) Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy->facexlib) (4.56.0) Requirement already satisfied: kiwisolver>=1.3.1 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy->facexlib) (1.4.8) Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy->facexlib) (24.1) Requirement already satisfied: pyparsing>=2.3.1 in /usr/lib/python3/dist-packages (from matplotlib->filterpy->facexlib) (2.4.7) Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy->facexlib) (2.9.0.post0) Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.11/dist-packages (from sympy->torch->facexlib) (1.3.0) Requirement already satisfied: six>=1.5 in /usr/lib/python3/dist-packages (from python-dateutil>=2.7->matplotlib->filterpy->facexlib) (1.16.0) Downloading facexlib-0.3.0-py3-none-any.whl (59 kB) Downloading numba-0.61.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (3.8 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.8/3.8 MB 196.7 MB/s eta 0:00:00 Downloading opencv_python-4.11.0.86-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (63.0 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 63.0/63.0 MB 115.5 MB/s eta 0:00:00 Downloading llvmlite-0.44.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (42.4 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.4/42.4 MB 163.4 MB/s eta 0:00:00 Building wheels for collected packages: filterpy Building wheel for filterpy (pyproject.toml) ... done Created wheel for filterpy: filename=filterpy-1.4.5-py3-none-any.whl size=110513 sha256=152dcc9b9d3e8b562126e3e347bdeb2a0d796e579f921ed9aae657262c1b556e Stored in directory: /root/.cache/pip/wheels/12/dc/3c/e12983eac132d00f82a20c6cbe7b42ce6e96190ef8fa2d15e1 Successfully built filterpy Installing collected packages: opencv-python, llvmlite, numba, filterpy, facexlib Successfully installed facexlib-0.3.0 filterpy-1.4.5 llvmlite-0.44.0 numba-0.61.0 opencv-python-4.11.0.86 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Collecting git+https://github.com/rodjjo/filterpy.git Cloning https://github.com/rodjjo/filterpy.git to /tmp/pip-req-build-s_ivm1de Running command git clone --filter=blob:none --quiet https://github.com/rodjjo/filterpy.git /tmp/pip-req-build-s_ivm1de Resolved https://github.com/rodjjo/filterpy.git to commit a34721e266c4c39f8abd5037ea83c2ed774d9606 Preparing metadata (setup.py) ... done Requirement already satisfied: matplotlib in /usr/local/lib/python3.11/dist-packages (from filterpy==1.4.5) (3.10.1) Requirement already satisfied: numpy in /usr/local/lib/python3.11/dist-packages (from filterpy==1.4.5) (1.26.3) Requirement already satisfied: scipy in /usr/local/lib/python3.11/dist-packages (from filterpy==1.4.5) (1.15.2) Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (1.3.1) Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (0.12.1) Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (4.56.0) Requirement already satisfied: kiwisolver>=1.3.1 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (1.4.8) Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (24.1) Requirement already satisfied: pillow>=8 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (10.2.0) Requirement already satisfied: pyparsing>=2.3.1 in /usr/lib/python3/dist-packages (from matplotlib->filterpy==1.4.5) (2.4.7) Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.11/dist-packages (from matplotlib->filterpy==1.4.5) (2.9.0.post0) Requirement already satisfied: six>=1.5 in /usr/lib/python3/dist-packages (from python-dateutil>=2.7->matplotlib->filterpy==1.4.5) (1.16.0) WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Collecting onnxruntime Downloading onnxruntime-1.21.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (4.5 kB) Collecting onnxruntime-gpu Downloading onnxruntime_gpu-1.21.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (4.8 kB) Collecting coloredlogs (from onnxruntime) Downloading coloredlogs-15.0.1-py2.py3-none-any.whl.metadata (12 kB) Collecting flatbuffers (from onnxruntime) Downloading flatbuffers-25.2.10-py2.py3-none-any.whl.metadata (875 bytes) Requirement already satisfied: numpy>=1.21.6 in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (1.26.3) Requirement already satisfied: packaging in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (24.1) Requirement already satisfied: protobuf in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (6.30.0) Requirement already satisfied: sympy in /usr/local/lib/python3.11/dist-packages (from onnxruntime) (1.12) Collecting humanfriendly>=9.1 (from coloredlogs->onnxruntime) Downloading humanfriendly-10.0-py2.py3-none-any.whl.metadata (9.2 kB) Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.11/dist-packages (from sympy->onnxruntime) (1.3.0) Downloading onnxruntime-1.21.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (16.0 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 16.0/16.0 MB 178.2 MB/s eta 0:00:00 Downloading onnxruntime_gpu-1.21.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (280.8 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 280.8/280.8 MB 140.3 MB/s eta 0:00:00 Downloading coloredlogs-15.0.1-py2.py3-none-any.whl (46 kB) Downloading flatbuffers-25.2.10-py2.py3-none-any.whl (30 kB) Downloading humanfriendly-10.0-py2.py3-none-any.whl (86 kB) Installing collected packages: flatbuffers, humanfriendly, coloredlogs, onnxruntime-gpu, onnxruntime Successfully installed coloredlogs-15.0.1 flatbuffers-25.2.10 humanfriendly-10.0 onnxruntime-1.21.0 onnxruntime-gpu-1.21.0 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Requirement already satisfied: opencv-python in /usr/local/lib/python3.11/dist-packages (4.11.0.86) Requirement already satisfied: numpy>=1.21.2 in /usr/local/lib/python3.11/dist-packages (from opencv-python) (1.26.3) WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Requirement already satisfied: opencv-python-headless in /usr/local/lib/python3.11/dist-packages (4.11.0.86) Requirement already satisfied: numpy>=1.21.2 in /usr/local/lib/python3.11/dist-packages (from opencv-python-headless) (1.26.3) WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Collecting ffmpeg-python Downloading ffmpeg_python-0.2.0-py3-none-any.whl.metadata (1.7 kB) Collecting future (from ffmpeg-python) Downloading future-1.0.0-py3-none-any.whl.metadata (4.0 kB) Downloading ffmpeg_python-0.2.0-py3-none-any.whl (25 kB) Downloading future-1.0.0-py3-none-any.whl (491 kB) Installing collected packages: future, ffmpeg-python Successfully installed ffmpeg-python-0.2.0 future-1.0.0 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Collecting ffmpeg Downloading ffmpeg-1.4.tar.gz (5.1 kB) Preparing metadata (setup.py) ... done Building wheels for collected packages: ffmpeg Building wheel for ffmpeg (setup.py) ... done Created wheel for ffmpeg: filename=ffmpeg-1.4-py3-none-any.whl size=6084 sha256=f0c97d48006fde387ba8336cbcded218aa2e0413d9cbd985d498908c03304138 Stored in directory: /root/.cache/pip/wheels/56/30/c5/576bdd729f3bc062d62a551be7fefd6ed2f761901568171e4e Successfully built ffmpeg Installing collected packages: ffmpeg Successfully installed ffmpeg-1.4 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'

Aitrepreneur

you're using the fp8 side, use the gguf side instead. for the other error send me the whole error log

Hometrees

hey sorry to bother so much but is there a way to fix this in runpod? Prompt outputs failed validation DualCLIPLoader: - Value not in list: clip_name2: 'llava_llama3_fp8_scaled.safetensors' not in [] - Value not in list: clip_name1: 'clip_l.safetensors' not in [] UNETLoader: - Value not in list: unet_name: 'hunyuan_video_FastVideo_720_fp8_e4m3fn.safetensors' not in [] VAELoader: - Value not in list: vae_name: 'hunyuan_video_vae_bf16.safetensors' not in ['sdxl_vae.safetensors'] I get this error when running the workflow now and am not too sue where to obtain the missing models ---------- Also when i run the .sh in run pod i am getting this ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt'

Hometrees

Omg thank you so much fixed

Aitrepreneur

using the template you get an older version of comfyui but then you won't be able to use the rest of the workflow

Aitrepreneur

if you used the updated installer you just need to re drag and drop the workflow inside comfyui again

Hometrees

Is there a way to install a previous version of comfyui on runpod so we do not have to worry about the bugs that come with updates?

Hometrees

i downloaded the updated version and still get this, is there a way for me to run previous version of comfy without auto updates to not have to worry about this?

Hometrees

I used the updated installer and got this error pls help DownloadAndLoadFlorence2Model (In group node 'workflow>Florence') workflow>Florence Florence2Run (In group node 'workflow>Florence') LoadFluxControlNet (In group node 'workflow>Flux Controlnet') workflow>Flux Controlnet ApplyFluxControlNet (In group node 'workflow>Flux Controlnet') XlabsSampler (In group node 'workflow>XLabs Sampler') workflow>XLabs Sampler ImpactGaussianBlurMask (In group node 'workflow>Nodes 3') workflow>Nodes 3 ReActorFaceBoost (In group node 'workflow>REACTORFACESWAP') workflow>REACTORFACESWAP ReActorFaceSwap (In group node 'workflow>REACTORFACESWAP') ReActorMaskHelper (In group node 'workflow>REACTORFACESWAP') SAMLoader (In group node 'workflow>Adetailer') workflow>Adetailer UltralyticsDetectorProvider (In group node 'workflow>Adetailer') FaceDetailer (In group node 'workflow>Adetailer') workflow>outpaintflor workflow>faceswap2 workflow>faceswap3 workflow>pullflo LongCLIPTextEncodeFlux (In group node 'workflow>pulidtxt') workflow>pulidtxt Text Concatenate (JPS) (In group node 'workflow>pulidtxt') PulidFluxInsightFaceLoader (In group node 'workflow>pluidlux') workflow>pluidlux PulidFluxEvaClipLoader (In group node 'workflow>pluidlux') PulidFluxModelLoader (In group node 'workflow>pluidlux') ApplyPulidFlux (In group node 'workflow>pluidlux') MaskBlur+ (In group node 'workflow>fillparam') workflow>fillparam workflow>PULIDREDUX Anything Everywhere (In group node 'workflow>REDUXOTHERPARAM') workflow>REDUXOTHERPARAM JWFloatSub (In group node 'workflow>Liveparam') workflow>Liveparam LTXAttnOverride (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTEXTOVIDGGUF ModifyLTXModel (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') LTXPerturbedAttention (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTXTFP workflow>LTXIMAGETOV workflowFlorence workflowFlux Controlnet workflowNodes 3 UltimateSDUpscale workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm

Aitrepreneur

This is some really weird issues, a combination or maybe some firewall/antivirus/restricted network or even outdated certificate.... So...you can try this: 1. Redownload this installer: https://www.patreon.com/file?h=121355091&m=431256208 2. Make sure you don't have any 7-zip already installed, if yes then uninstall it. If you don't have 7-zip installed already. Then install 7z in your machine via the site installer: https://www.7-zip.org/a/7z2201-x64.exe then install it in C:\Program Files by default. Then add 7zip to path, for this go into the Properties → Advanced system settings → Environment Variables Edit the "Path" variable under System variables set the path in Environment Variables -> User Variables path -> edit -> new (add the path - C:\Program Files\7-Zip) -> OK Close all windows. 3. Relaunch the installer

Juan Forero

In all the things I try to install, I get the same issues: Hello, I get this message: [INFO] Installing vs_BuildTools... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 curl: (35) schannel: next InitializeSecurityContext failed: CRYPT_E_NO_REVOCATION_CHECK (0x80092012) -The revocation function cannot check the revocation status for the certificate [ERROR] Download failed. Please restart the installer Then it continues and I get this error: 7-Zip is not installed. Downloading and installing... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 154 100 154 0 0 282 0 --:--:-- --:--:-- --:--:-- 282 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 100 1538k 100 1538k 0 0 1295k 0 0:00:01 0:00:01 --:--:-- 48.4M Installation of 7-Zip failed. Please install it manually and try again. I install mannually 7-zip and get the same error. And then, when it continues, I get this: Git already installed. Downloading ComfyUI... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 100 1604M 100 1604M 0 0 43.7M 0 0:00:36 0:00:36 --:--:-- 42.3M Extracting ComfyUI... El sistema no puede encontrar la ruta especificada. (the system cannot find the specific path) Extraction failed. Please check the downloaded file and try again.

Elrond

Here's the new one. Install custom node 'ComfyUI Impact Pack' Install: ['https://github.com/ltdrdata/ComfyUI-Impact-Pack'] Download: git clone 'https://github.com/ltdrdata/ComfyUI-Impact-Pack' [!] 100%|██████████| 147.0/147.0 [00:01<00:00, 107.35it/s] 100%|██████████| 147.0/147.0 [00:01<00:00, 107.35it/s] 100%|██████████| 34.0/34.0 [00:01<00:00, 24.80it/s] 100%|██████████| 34.0/34.0 [00:01<00:00, 24.80it/s] 87%|████████▌ | 4672.0/5392.0 [01:47<00:16, 43.59it/s]Cmd('git') failed due to: exit code(128) 87%|████████▌ | 4672.0/5392.0 [01:47<00:16, 43.57it/s] cmdline: git clone -v --recursive --progress -- https://github.com/ltdrdata/ComfyUI-Impact-Pack D:\wan\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack

Elrond

And that. [INFO] Checking Python packages... [INFO] Determining Python version... [INFO] Detected Python version: 3.12 [SKIP] InsightFace wheel file already exists. [INFO] Installing required Python packages... Collecting facexlib Using cached facexlib-0.3.0-py3-none-any.whl.metadata (4.6 kB) Collecting filterpy (from facexlib) Using cached filterpy-1.4.5-py3-none-any.whl ERROR: Could not install packages due to an OSError: [Errno 13] Permission denied: 'C:\\users\\admin\\appdata\\local\\pip\\cache\\wheels\\77\\bf\\4c\\b0c3f4798a0166668752312a67118b27a3cd341e13ac0ae6ee\\filterpy-1.4.5-py3-none-any.whl' Consider using the `--user` option or check the permissions.

Elrond

I am getting lots of errors while installing with bat file. I've been trying to install for like 48 hours, tried to download maybe 5-6 times (lost count on that really) but no luck yet. Where are those pre-downloaded files stored? I am suspecting that maybe one or few of those are broken and that's why messing with install. Latest errors I get are those: [INFO] Setting up custom nodes repositories... [INFO] Checking for ComfyUI-Manager... [INFO] Cloning ComfyUI-Manager repository... Cloning into 'ComfyUI-Manager'... remote: Enumerating objects: 18185, done. remote: Counting objects: 100% (2966/2966), done. remote: Compressing objects: 100% (244/244), done. error: RPC failed; curl 92 HTTP/2 stream 5 was not closed cleanly: CANCEL (err 8) error: 51527 bytes of body are still expected fetch-pack: unexpected disconnect while reading sideband packet fatal: early EOF fatal: fetch-pack: invalid index-pack output --------------------------

Kusacubari

seemed to fix further probs >> run cmd in folder where files are to convert the .pth files to .safetensors import torch from safetensors.torch import save_file # Load the .pth file model = torch.load("appearance_feature_extractor.pth", map_location="cpu") # Convert and save it as a .safetensors file save_file(model, "appearance_feature_extractor.safetensors")

Kusacubari

turns out there was a quick fix... the files were called .pth which is why it wasn't finding the file... gpt said try renaming them to .safetensors and it worked.

Kusacubari

wondered if there is an easy answer to this error for the LivePortrait (both manual and v2v) DownloadAndLoadLivePortraitModels No such file or directory: "V:\\ComfyUI-V3AIO\\ComfyUI\\models\\liveportrait\\appearance_feature_extractor.safetensors" the files are there (although path seems to miss out the "base_models" dir, but even when I put the file its looking for where it thinks it is (according to the path) it still does work and throws this error. thanks.

Aitrepreneur

use the newest updated installer

jonescrusher1

I've followed instructions for runpod installation and unable to get past these missing nodes: When loading the graph, the following node types were not found: workflow>LTXPARAMTEXTOVIDGGUF workflow>LTXPARAMTXTFP workflow>LTXIMAGETOV workflow>wanvaeim workflow>wanparam workflow>wanvae workflow>skyreelsparam workflow>skyreelslogic workflow>skyreelsotherparam Label (rgthree) workflow>WANIMAGEPARAM workflow>hynupar Fast Groups Bypasser (rgthree)

Aitrepreneur

I can't try it out myself so I can't be 100% sure indeed, would be great to have more people's trying

Junbi

It will, but you need to install a different version of ComfyUI and copy the files over manually from the Auto Installer. Not sure why I seem to be the only one letting people know this. I know there aren't a ton of 50XX cards, but there are enough that I think it should be mentioned in the install description at least. 50XX ComfyUI: https://github.com/comfyanonymous/ComfyUI/discussions/6643

Aitrepreneur

send me a dm

Aitrepreneur

Link to the workflow is in the post

Aitrepreneur

Download the updated installer

Aitrepreneur

I updated the installer, it should work now

Aitrepreneur

Just added the new bat file for that

David Alkhoutoff

Does this work for the 50 series?

Anton Karpuzikov

error: unable to create file src/custom_controlnet_aux/depth_anything/torchhub/facebookresearch_dinov2_main/dinov2/configs/eval/vitb14_pretrain.yaml: Filename too long error: unable to create file src/custom_controlnet_aux/depth_anything/torchhub/facebookresearch_dinov2_main/dinov2/configs/eval/vitg14_pretrain.yaml: Filename too long error: unable to create file src/custom_controlnet_aux/depth_anything/torchhub/facebookresearch_dinov2_main/dinov2/configs/eval/vitl14_pretrain.yaml: Filename too long error: unable to create file src/custom_controlnet_aux/depth_anything/torchhub/facebookresearch_dinov2_main/dinov2/configs/eval/vits14_pretrain.yaml: Filename too long fatal: unable to checkout working tree warning: Clone succeeded, but checkout failed. You can inspect what was checked out with 'git status' and retry with 'git restore --source=HEAD :/' Failed to clone repository.

Memo Testi

can i download just the workflow please. i want to do the rest manually :-) where do i find it?

Sena

This didn't work for me unfortunately.

avoN_C

Yeah I was expecting my existing installation to be "updated" but instead it redownloads the whole thing again. *sigh

Sena

Yup same here

Taiga

Ran the v3 bat in an empty directory, in a new windows user profile (that "fixed" v2)... still just a mess: Missing Node Types When loading the graph, the following node types were not found DownloadAndLoadFlorence2Model (In group node 'workflow>Florence') workflow>Florence Florence2Run (In group node 'workflow>Florence') LoadFluxControlNet (In group node 'workflow>Flux Controlnet') workflow>Flux Controlnet ApplyFluxControlNet (In group node 'workflow>Flux Controlnet') XlabsSampler (In group node 'workflow>XLabs Sampler') workflow>XLabs Sampler ImpactGaussianBlurMask (In group node 'workflow>Nodes 3') workflow>Nodes 3 ReActorFaceBoost (In group node 'workflow>REACTORFACESWAP') workflow>REACTORFACESWAP ReActorFaceSwap (In group node 'workflow>REACTORFACESWAP') ReActorMaskHelper (In group node 'workflow>REACTORFACESWAP') SAMLoader (In group node 'workflow>Adetailer') workflow>Adetailer UltralyticsDetectorProvider (In group node 'workflow>Adetailer') FaceDetailer (In group node 'workflow>Adetailer') workflow>outpaintflor workflow>faceswap2 workflow>faceswap3 workflow>pullflo LongCLIPTextEncodeFlux (In group node 'workflow>pulidtxt') workflow>pulidtxt PulidFluxInsightFaceLoader (In group node 'workflow>pluidlux') workflow>pluidlux PulidFluxEvaClipLoader (In group node 'workflow>pluidlux') PulidFluxModelLoader (In group node 'workflow>pluidlux') ApplyPulidFlux (In group node 'workflow>pluidlux') MaskBlur+ (In group node 'workflow>fillparam') workflow>fillparam workflow>PULIDREDUX Anything Everywhere (In group node 'workflow>REDUXOTHERPARAM') workflow>REDUXOTHERPARAM JWFloatSub (In group node 'workflow>Liveparam') workflow>Liveparam LTXAttnOverride (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTEXTOVIDGGUF ModifyLTXModel (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') LTXPerturbedAttention (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTXTFP workflow>LTXIMAGETOV workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowLiveparam workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV

Cheg

FLUX LORA TRAINING + LORA TESTING WORKFLOW

Taiga

same

Taiga

same. as usual, I got tricked into thinking that this would work.

Michael

This would be awesome.

Taiga

You mean in the fluxgym post? I'm not seeing it, ffs

Adictedtohumping

can you just make a bat file with the model downloads? some run one stability matrix and dont want to download comyui for the 200th time ;)

Aitrepreneur

I'm aware of the issue, it's because of a recent comfyui update that broke some stuff, I will fix the installer asap no worries!

Cheg

same here

Cheg

it is in the flux lora training post

Omer Nachshon

Thank you @Dani. You're a true hero.

Nathan Ternan

where is it though?

Omer Nachshon

update: got some working by manually installing the missing nodes. still cant find a way to fix these: LTXAttnOverride (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTEXTOVIDGGUF ModifyLTXModel (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') LTXPerturbedAttention (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTXTFP workflow>LTXIMAGETOV workflowpulidtxt workflowREDUXOTHERPARAM workflowLiveparam workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV

EX01

Same as me, just cannot get rid no matter what I do?

ES

Same issue for me too. Only V2 when it comes to clicking queue works but not V3 or V3.1

ES

Same issue

Omer Nachshon

did everything, same as v2, still getting n loading the graph, the following node types were not found DownloadAndLoadFlorence2Model (In group node 'workflow>Florence') workflow>Florence Florence2Run (In group node 'workflow>Florence') LoadFluxControlNet (In group node 'workflow>Flux Controlnet') workflow>Flux Controlnet ApplyFluxControlNet (In group node 'workflow>Flux Controlnet') XlabsSampler (In group node 'workflow>XLabs Sampler') workflow>XLabs Sampler ImpactGaussianBlurMask (In group node 'workflow>Nodes 3') workflow>Nodes 3 ReActorFaceBoost (In group node 'workflow>REACTORFACESWAP') workflow>REACTORFACESWAP ReActorFaceSwap (In group node 'workflow>REACTORFACESWAP') ReActorMaskHelper (In group node 'workflow>REACTORFACESWAP') SAMLoader (In group node 'workflow>Adetailer') workflow>Adetailer UltralyticsDetectorProvider (In group node 'workflow>Adetailer') FaceDetailer (In group node 'workflow>Adetailer') workflow>outpaintflor workflow>faceswap2 workflow>faceswap3 workflow>pullflo LongCLIPTextEncodeFlux (In group node 'workflow>pulidtxt') workflow>pulidtxt Text Concatenate (JPS) (In group node 'workflow>pulidtxt') PulidFluxInsightFaceLoader (In group node 'workflow>pluidlux') workflow>pluidlux PulidFluxEvaClipLoader (In group node 'workflow>pluidlux') PulidFluxModelLoader (In group node 'workflow>pluidlux') ApplyPulidFlux (In group node 'workflow>pluidlux') MaskBlur+ (In group node 'workflow>fillparam') workflow>fillparam workflow>PULIDREDUX Anything Everywhere (In group node 'workflow>REDUXOTHERPARAM') workflow>REDUXOTHERPARAM JWFloatSub (In group node 'workflow>Liveparam') workflow>Liveparam LTXAttnOverride (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTEXTOVIDGGUF ModifyLTXModel (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') LTXPerturbedAttention (In group node 'workflow>LTXPARAMTEXTOVIDGGUF') workflow>LTXPARAMTXTFP workflow>LTXIMAGETOV workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm

ZZZ0mbieSSS

Have you fixed it? For me it showing in red only the combined nodes

Jason Blake

This fixed everything went smooth after doing this on Windows 11, Portable Python environmental path was going to my appdata, to avoid this do this:>>>> Go to -> "start" and type "Manage App Execution Aliases". Go to it and turn off "Python"<<<<<

Ez

Everything works great, but I have a couple of small issues. "Error. No naistyles.csv found. Put your naistyles.csv in the custom_nodes/ComfyUI_NAI-mod/CSV directory of ComfyUI. Then press "Refresh"." I did that, but I still get the error. I don't even know what naistyles does. I had the same issue with the older versions as well. It doesn't seem to affect anything though... Should I care? Also, the controlnet workflows seem to add their own model info from somewhere. It makes them unusable since, even though it matches the pose perfectly, the output isn't the model/lora combination I want. Is this a settings thing? It happens even with the default settings. The in/outpainting has a similar, though far weaker, effect. I do primarily create anime and manga style images though, so perhaps that's why?

Kevin Vandendriessche

I have the same issue

Tbatz

LMK if this gets fixed for you

Tbatz

Same

Martin T

Thanks, managed to get it sorted

Omer Nachshon

same problem

Papalap

never mind im an idiot xD

Papalap

where do i find the "V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json file"????

Thomas Collard

Thanks a lot @Dani, you fixed the issue for me !

Kubilay göze

Actually I solved this in the weirdest way so I don't think it will help anyone but here it goes: I had my previous ComfyUI_windows_portable folder as a backup when I installed this one from scratch. I copied the new ComfyUI_windows_portable folder again as a 3rd copy (so I would have backups for both) I copied my old ComfyUI (the one inside ComfyUI_windows_portable) folder into the new one, overwritten the same documents. I don't know why I had this idea but it somehow worked. Some of the new files and old files got along better I guess. See what I come to realize with open source softwares is that it's more art than science :D

Dani

I think you will need to drag and drop your workflow JSON once again into the comfyui window and it will fix those errors.

EX01

So despite installing all the custom nodes and updating them and dragging and dropping the workflow back onto and tried using the load workflow again, still getting missing nodes red boxes.

Martin T

I managed to install most of the nodes using your method, but what about the workflow? How do you install the missing workflow such as workflowREDUXOTHERPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV

J2

Um... I might be missing something. I seem to have everything installed correctly. Everything appears to work ok until I click queue. Then nothing happens. It doesn't matter which model I am using, whether it's the FP8 or Hunyuan, or gguf. Nothing happens when I click queue

Dani

This is what I did to fix all the missing nodes (Had about the same amount of missing nodes as you): Note down the name of the missing node(s) (for example "applyfluxcontrolnet") . Go to https://ltdrdata.github.io/ and search for the noted down node name. You will find the nodes name in the "Title" column you will need to install. For applyfluxcontrolnet it is "x-flux-comfyui". Head back to ComfyUI > Manager and choose "Custom Nodes Manager". Paste the name (in this example "x-flux-comfyui") in the searchbox and click on install. Then click on Restart and wait for ComfyUI to reload. Following these steps for all the missing nodes I was able to finally start ComfyUI without any issue. There was one node that was tricky, called "Anything Everywhere", for this you will need to install "cg-use-everywhere".

Aitrepreneur

send me a dm

Aitrepreneur

send me a dm

Crypto Funky

just not working

GO Zurn

Same problem here too

ZZZ0mbieSSS

Same for me.

ES

Same issue for me too

Crypto Funky

Non funziona nemmeno su RUNPOD, bel pacco! Grande

Kevin Vandendriessche

how do i get this to work... i try something but nothing happens when pressing queue

Kubilay göze

I used the installer from scratch even though I had already a comfyui installed. (I changed the name with "OLD") the install seemed to be went fine. I installed all the missing custom nodes now I get the "following node types not found" florence, applyfluxcontrolnet, xlabssampler, reactorfaceswap etc... around 15-20 nodes on the list. "Install missing custom nodes" is completely empty AND I DID DRAG THE WORKFLOW AGAIN so please nobody give me that advice again. I did all the updates, the refreshing, all the things in the video and the troubleshoot video. what do I do? I downloaded 250gb of data and now I'm just sitting here unable to play with it, it's kinda frustrating...

Crypto Funky

Anche io ho lo stesso problema, installato tutto (spero, 160gb circa di cartella), aggiornato tutto e installato i nodi mancanti ma quando provo a generare non succede assolutamente nulla, molti box sono ancora rossi e non riesco a capire perchè... Se qualcuno avesse un idea sarebbe fantastico. Grazie.

Lionydus

I fixed this by going to V3 GGUF REDUX + PULID and FP8 REDUX + PULID and deleting REDUXOTHERPARAM node in the top middle. It's sticky. Have to delete them multiple times. Press delete. Right click remove. I don't know how to fix those bugged nodes. Been trying for hours.

Chris Spears

Installed the one click auto install twice, and did the manual install once, and all 3 times, once I get into comfyUI, I can not use any of the nodes. I try to do a basic image generation on gguf side, and nothing happens in the terminal. I changed the parameters and seed and a different prompt, and still nothing. Some nodes still have red boxes for missing items, but my terminal shows everything is ok, no errors. Nothing shows up on install missing nodes, and COmdy is updated to latest. Any ideas why I can't even generate an image?

BadgerWi Johnson

I was a lot. We spent about an hour working through each issue that popped up and fixing it individually. All I did was copy and paste the error that showed up in the CMD box into Copilot and it would tell me how to fix that error.

Kevin Vandendriessche

the installer just stops.... and nothing happens, great money spend

brentwd

what did copilot say? im having same issues

Gavin Lister

Probably your best work yet! I've been playing around with flux loras from civitai using the workflow and have been getting outputs that are photolike with just 1 pass (1x 2 minute generation on a 3080ti, default settings 25 steps/Euler). Worth every penny of the Subscription!

JpaLVVmI6c8xM00

Would it be possible to make a 1 click installer for amd cards? I am having issues getting things working, and I suspect it's because I am trying cpu only.

Jason Blake

Having the same issue, everything wants to path to C:\Users\UserName\appdata\ how can I point everything to C:\Users\UserData\AI\COMFY

Daniel Kutz

I asked run pod to help fix the problem, involving some fix of comfyui. This is what the tech guy did and it fixed it (don't know why it happens in the first place) - So the first issue I fiexed by activating Comfy UI vene with command: source /workspace/ComfyUI/venv/bin/activate Then inside of it I installed missing package with: pip install propcache Then I fixed other custom nodes with: comfy node restore-dependencies Is there anything else I can help you with or it's fine to close ticket for now?

Daniel Kutz

Happened to me also three times already. Something is weird.

arif hidayat

hello, can we use secure cloud in runpod to use this workflow

saiprasad P

not getting comfy manager

saiprasad P

E:\comfyui\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build [START] Security scan [DONE] Security scan ## ComfyUI-Manager: installing dependencies. (GitPython) [notice] A new release of pip is available: 24.3.1 -> 25.0.1 [notice] To update, run: E:\comfyui\ComfyUI_windows_portable\python_embeded\python.exe -m pip install --upgrade pip ## ComfyUI-Manager: installing dependencies done. ** ComfyUI startup time: 2025-02-24 07:33:47.721 ** Platform: Windows ** Python version: 3.12.8 (tags/v3.12.8:2dc476b, Dec 3 2024, 19:30:04) [MSC v.1942 64 bit (AMD64)] ** Python executable: E:\comfyui\ComfyUI_windows_portable\python_embeded\python.exe ** ComfyUI Path: E:\comfyui\ComfyUI_windows_portable\ComfyUI ** ComfyUI Base Folder Path: E:\comfyui\ComfyUI_windows_portable\ComfyUI ** User directory: E:\comfyui\ComfyUI_windows_portable\ComfyUI\user ** ComfyUI-Manager config path: E:\comfyui\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini ** Log path: E:\comfyui\ComfyUI_windows_portable\ComfyUI\user\comfyui.log WARNING: The script f2py.exe is installed in 'E:\comfyui\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. gradio 5.8.0 requires ffmpy, which is not installed. omnigen 1.0.3 requires accelerate==0.26.1, which is not installed. peft 0.14.0 requires accelerate>=0.21.0, which is not installed. [notice] A new release of pip is available: 24.3.1 -> 25.0.1 [notice] To update, run: E:\comfyui\ComfyUI_windows_portable\python_embeded\python.exe -m pip install --upgrade pip Prestartup times for custom nodes: 21.8 seconds: E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager Checkpoint files will always be loaded safely. Total VRAM 6144 MB, total RAM 40661 MB pytorch version: 2.6.0+cu126 Set vram state to: NORMAL_VRAM Device: cuda:0 NVIDIA GeForce RTX 3060 Laptop GPU : cudaMallocAsync Using pytorch attention ComfyUI version: 0.3.15 [Prompt Server] web root: E:\comfyui\ComfyUI_windows_portable\ComfyUI\web Traceback (most recent call last): File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2145, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait\__init__.py", line 1, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait\nodes.py", line 11, in import dill ModuleNotFoundError: No module named 'dill' Cannot import E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait module for custom nodes: No module named 'dill' Traceback (most recent call last): File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2145, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\__init__.py", line 1, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\nodes.py", line 18, in from .liveportrait.live_portrait_pipeline import LivePortraitPipeline File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\liveportrait\live_portrait_pipeline.py", line 16, in from .utils.filter import smooth File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\liveportrait\utils\filter.py", line 3, in from pykalman import KalmanFilter ModuleNotFoundError: No module named 'pykalman' Cannot import E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ module for custom nodes: No module named 'pykalman' Traceback (most recent call last): File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2145, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\__init__.py", line 8, in import manager_server # noqa: F401 ^^^^^^^^^^^^^^^^^^^^^ File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\glob\manager_server.py", line 17, in import manager_core as core File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\glob\manager_core.py", line 29, in from rich import print ModuleNotFoundError: No module named 'rich' Cannot import E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager module for custom nodes: No module named 'rich' Traceback (most recent call last): File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2145, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node\__init__.py", line 23, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node\nodes.py", line 15, in from insightface.app.common import Face File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\insightface\__init__.py", line 18, in from . import app File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\insightface\app\__init__.py", line 2, in from .mask_renderer import * File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\insightface\app\mask_renderer.py", line 4,in import albumentations as A File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\albumentations\__init__.py", line 18, in from .augmentations import * File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\albumentations\augmentations\__init__.py",line 1, in from .blur.functional import * File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\albumentations\augmentations\blur\__init__.py", line 1, in from .functional import * File "E:\comfyui\ComfyUI_windows_portable\python_embeded\Lib\site-packages\albumentations\augmentations\blur\functional.py", line 13, in from pydantic import ValidationInfo ModuleNotFoundError: No module named 'pydantic' Cannot import E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node module for custom nodes: Nomodule named 'pydantic' Import times for custom nodes: 0.0 seconds: E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py 0.0 seconds (IMPORT FAILED): E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager 0.0 seconds (IMPORT FAILED): E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ 0.0 seconds: E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale 0.0 seconds (IMPORT FAILED): E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait 0.1 seconds (IMPORT FAILED): E:\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node Starting server To see the GUI go to: http://127.0.0.1:8188

Maurício Velasquez

235 / 5.000 All tools give this error: LTXVImgToVideo The expanded size of the tensor (24) must match the existing size (96) at non-singleton dimension 4. Target sizes: [1, 128, 4, 16, 24]. Tensor sizes: [4, 64, 96] I'm running via Runpod.

arif hidayat

1 person faceswap error : # ComfyUI Error Report ## Error Details - **Node ID:** 1971 - **Node Type:** ReActorMaskHelper - **Exception Type:** _pickle.UnpicklingError - **Exception Message:** Weights only load failed. This file can still be loaded, to do so you have two options, ?[1mdo those steps only if you trust the source of the checkpoint?[0m. (1) In PyTorch 2.6, we changed the default value of the `weights_only` argument in `torch.load` from `False` to `True`. Re-running `torch.load` with `weights_only` set to `False` will likely succeed, but it can result in arbitrary code execution. Do it only if you got the file from a trusted source. (2) Alternatively, to load with `weights_only=True` please check the recommended steps in the following error message. WeightsUnpickler error: Unsupported global: GLOBAL ultralytics.nn.tasks.DetectionModel was not an allowed global by default. Please use `torch.serialization.add_safe_globals([DetectionModel])` or the `torch.serialization.safe_globals([DetectionModel])` context manager to allowlist this global if you trust this class/function.

Dragonbait

you need to go into the ComfyUI folder and drag the .sh file there, not just the root folder

Karam El Wardi

For some reason it said : ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt' See full error below. Any help please ? [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip Collecting ffmpeg Downloading ffmpeg-1.4.tar.gz (5.1 kB) Preparing metadata (setup.py) ... done Building wheels for collected packages: ffmpeg Building wheel for ffmpeg (setup.py) ... done Created wheel for ffmpeg: filename=ffmpeg-1.4-py3-none-any.whl size=6082 sha256=4c72f673c3090cc29362d135de630a31e880475f8d34c2004526824a7b47b8eb Stored in directory: /root/.cache/pip/wheels/8e/7a/69/cd6aeb83b126a7f04cbe7c9d929028dc52a6e7d525ff56003a Successfully built ffmpeg Installing collected packages: ffmpeg Successfully installed ffmpeg-1.4 WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable.It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip [notice] A new release of pip is available: 24.2 -> 25.0.1 [notice] To update, run: python3 -m pip install --upgrade pip ERROR: Could not open requirements file: [Errno 2] No such file or directory: 'requirements.txt' # #

ahazy

Nobody replied to help me so I'm buying a new SSD and windows license and redownloading for the third time. Maybe it worked, maybe it didn't.

lolno

Does anyone has this issue when trying to run image to video models? ModifyLTXModel 'LTXVModelModified' object has no attribute 'transformer_blocks'

Cameron Tavassoli

I fixed it by deleting everything and then reinstalling clean and following the troubleshooting video. Thanks.

Cameron Tavassoli

I messed something up, can you make a one-click-ultimate all flux models and comfy ui uninstaller too? thanks.

Jean Dupont

Same for me. And i don't have the start of a clue on how to apply the solution from Bil ^^'

ahazy

Will this install work on an external SSD? Not C: drive.

Cinemalecular

I'm sure my installation wasn't recent, although I've kept git up to date with TortoiseGit. It may be one of those things that the newer binary installations have it included. Maybe just something to add to the readme. Thanks.

BadgerWi Johnson

I had MS Copilot talk me through the errors and I reloaded it and we got it going. All hail our Ai overlords! 😀

Aitrepreneur

Are you sure you weren't using an older version of git? Because right now git lsf is included in the windows distribution of git by default (look here: https://github.com/git-lfs/git-lfs?tab=readme-ov-file#on-windows and here's the proof here: https://imgur.com/a/WksUmt0 it's checked by default during the install). You might have just been using an older version of git where it wasn't installed by default.

Aitrepreneur

check the troubleshooting video if it's your first time using a comfyui workflow: https://www.patreon.com/posts/116007628

Express Appel

Hey! I have a small issue on runpod version, everything was fine, then i clicked on "Stop" to close the pod, i tried to launch it again and now the port 3000 show me this error "The port is not up yet". How can i start comfyui from the console please ?

Jordan Hewitt

did you ever figure this out? because this is also happening for me

Bil Mor

In case its helpful, I solved this for myself (well, mostly Github Copilot solved it) by putting the looping retries inside this function at the top of the file, and then calling it throughout the file rather than using gotos: @echo off setlocal enabledelayedexpansion goto :main :: === Function Definitions === :download_file set "attempt=1" set "max_attempts=3" set "file_path=%1" set "url=%2" :retry_download echo Attempt %attempt% of %max_attempts% to download %file_path%... curl -L -C - -o %file_path% %url% --ssl-no-revoke --connect-timeout 60 --max-time 3600 --retry 3 --retry-delay 10 if %errorlevel% NEQ 0 ( echo Download attempt %attempt% failed. set /a "attempt+=1" if %attempt% LEQ %max_attempts% ( echo Retrying download in 30 seconds... timeout /t 30 /nobreak goto retry_download ) else ( echo All download attempts failed for %file_path%. choice /C YN /M "Do you want to continue with the installation without this file" if errorlevel 2 exit /b 1 echo Continuing installation... ) ) else ( echo Successfully downloaded %file_path% ) exit /b 0 :: === END OF Function Definitions === :main REM Environment Variables winget set "winget_path=%userprofile%\AppData\Local\Microsoft\WindowsApps" EXAMPLE OF A CALL (Not the most elegant, but copilot's fix worked, so I won't complain): :: === Download with retry and resume for t5xxl_fp8_e4m3fn.safetensors === if exist "t5xxl_fp8_e4m3fn.safetensors" ( choice /C YN /M "t5xxl_fp8_e4m3fn.safetensors already exists. Download again" if errorlevel 2 ( echo Skipping t5xxl_fp8_e4m3fn.safetensors download... ) else ( del "t5xxl_fp8_e4m3fn.safetensors" call :download_file "t5xxl_fp8_e4m3fn.safetensors" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/t5xxl_fp8_e4m3fn.safetensors" ) ) else ( call :download_file "t5xxl_fp8_e4m3fn.safetensors" "https://huggingface.co/Aitrepreneur/FLX/resolve/main/t5xxl_fp8_e4m3fn.safetensors" )

Bil Mor

The problem is that these scripts execute goto's to a label inside of an if/else block (such as :retry_t5xxl_fp16). Batch files do NOT like that. It ends up executing the else statement after the if statement, which puts it in a loop.

Bil Mor

I tried removing that entry on the script, and it just starts repeating the next model.

Jean Bagarre

same as you

Rallyace

I have exactly the same problem and reloading the workflow did not fix it. Any help would be appreciated.

Firmy Van Bas

How big is the final folder supposed to be? I ran out of space during the install and the folder is already 300gb.

Bil Mor

Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1138 100 1138 0 0 4506 0 --:--:-- --:--:-- --:--:-- 4533 100 49 100 49 0 0 162 0 --:--:-- --:--:-- --:--:-- 162 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1138 100 1138 0 0 8774 0 --:--:-- --:--:-- --:--:-- 8890 100 49 100 49 0 0 279 0 --:--:-- --:--:-- --:--:-- 279 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1134 100 1134 0 0 10316 0 --:--:-- --:--:-- --:--:-- 10500 100 49 100 49 0 0 324 0 --:--:-- --:--:-- --:--:-- 324 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors

Bil Mor

my one click installer gets stuck in a loop downloading t5xxl_fp8_e4m3fn.safetensors over and over again.

Cinemalecular

If you're referring to line 321, that line fails unless you've already installed the git-lfs.exe (https://git-lfs.com). It's add-on command for git that isn't installed by default. the git lfs install command only seeks to enable for the local repo...but it requires the exe to be added to the git bin first.

DJ

Can I install this with pinokio?

James Pratt

Are there any scripts for installing in OSX? Linux .sh failed.

Aitrepreneur

yes it's already in the installer, it does it whenever it needs to clone bit file repositories

Express Appel

Okay, i got this error : Installation of 7-Zip failed. Please install it manually and try again. I already had 7Zip installed, i still tried to use the 7Zip installer downloaded by the .bat but still have this error and nothing else happen EDIT : Found a solution with chatgpt, i needed to add the path for 7z manually, i'm still in the installation process right now, but i don't have this issue anymore :)

Cinemalecular

This may have already been discovered in other versions, but you may need to install git-lfs and initialize it. That was one of the problems I ran into. That allows for the download of larger files from git repositories and it isn't installed with git by default.

Aitrepreneur

you need to use the V3 installer from scratch since it uses the brand new comfyui torch version

Aitrepreneur

don't launch as admin, try opening a new cmd window and drag and dropping the installer inside that window then press enter, this will prevent the window from closing and at least gives us the explanation of why it closes

Aitrepreneur

yes but you need the special comfyui for amd gpu: https://github.com/patientx/ComfyUI-Zluda/pulls

Mourad

Reload the workflow and it will be fixed.

gabbie Lopez

if we already downloaded the v2 bat file do we have to download from scratch the v3, or can we just download everything through the comfyui manager?

BadgerWi Johnson

Ok new problem finally got the manager up and running, boots with no errors but when ComfyUI launches I am met with a list of nodes that didn't load. I have updated the UI and here are no nodes in the node list to be updated. I try generating a simple image and nothing happens so these nodes must be important. :) . Any thoughts? When loading the graph, the following node types were not found workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm workflowREACTORFACESWAP

BadgerWi Johnson

I entered all my errors into Microsoft co-pilot and it took a while but eventually it got it all straight and the manager works I am currently updating the nodes as I write this.

K1mo7chi

Its Visual Studio (had same problem but also installing it havent fixed my manager)

K1mo7chi

Same here installing manager from cmd don't work :(

Express Appel

I don't really understand why but when i try to launch the installer, i get a 7z install screen, if i click yes, everything close, if i click no, everything close too EDIT : If i launch the installer in admin mode, it closes even before when VS installer is launched

Mark Long

Will it work with AMD GPU? 7900 XT etc?

Caine Ruiz

same issue, (installer says installing git and then it does its thing for a while but closes out shortly after and nothing happens)

Occavango

Does the faceswap connect to the internet? I’ve blocked web access and I get an error when I try and run faceswap, “urlopen error [WinError 10013] An attempt was made to access a socket in a way forbidden by its access permissions”. Not overly keen for my nsfw face swaps to be routing via the internet

Dreddoom

Hi, I've installed the ultimate workflow. But there is no workflow file. Is it on here somewhere? Thanks Nevermind, found it in the Lara training link 😵‍💫

Thomas

Anyone having issues with loras ? It seems loras don't load when toggled on. Also, can't switch out checkpoints.

MPG

I typically like using your installers because they're so dependable but I've been having way more issues than I'd like with this one 1. after I do the initial prompts I get an error " cannot open app package. Reason: The package is either corrupted or invalid. 2. (this might be a byproduct of the first error) absolutely nothing is being downloaded with the installer besides comfy. No Models, custom nodes, or anything. I don't know what to do. Would really appreciate some help

Explore Hub

So i installed the V3 ultimate flux comfui manager auto, and nothing happened. Nor could i find the file.

E Farts

Since I don't see anyone mentioning the Manager except one person down below, I had to go to the ComfyUI website, search for the manager files, and follow the instructions to install it manually. (Went to the file location and typed into the location bar "CMD" to bring up the terminal at that location. Then ran the code it said to. "git clone https://github.com/ltdrdata/ComfyUI-Manager.git"). After that I ran the ComfyUI update bat for good measure and when I started it up, it was there in the top right. Hope this helps!(https://docs.comfy.org/essentials/core-concepts/nodes#comfyui-manager)

Aitrepreneur

send me a dm

VladoPortos

Same issue, no matter what I do I can't get the manager in confui and therefore I can't install any missing nodes :(

Vilain le Singe

Inpainting and Redux doens't work, i barely change the image - it doesn't use the prompt AT ALL, any help ? :D

Alexandre Graça

Hi Bradford. Would you mind sharing links for the 90GB Hunyuan and how to merge with Aitrepreneur brew?

toxrednil

Thank you for adding the .sh versioin of the V3 Install file! Yay!

Vilain le Singe

Weirdly, Inpainting (and REDUX)doesn't work. It just don't use the prompt, it just redux my existing picture. Blocking the seed and changing prompt create the same image over and over. If i change the model from fill to dev while keeping the denoise at 1.0, the masked part is redux hard, but still not using one word of the prompt :0 Help me Overlord !

Bunny Swallows

Hey, glad to see you back and hope that everything stays on track for you man. thank you for the new video and workflow too!

LW

You can always load compatible models in to the unet folder and then select them from the relevant workflow parameter box. Note that if ComfyUI is already running you'll need to restart or reload the page so that the V3 workflow knows that a new model is available.

Thomas

I had this problem, it downloads Git but then i had to manually install it. I had problems down the line, not sure if its related.

Thomas

I'm having some issues installing this even after watching the troubleshooting video. I am getting this error when trying to make a video. LTXVModelModified' object has no attribute 'transformer_blocks - i have uninstalled and reinstalled several times and still having an issue.

Occavango

Noob question, but can I play around with other models in this workflow? I have a heap of checkpoints that I love, can I use those? Can I use Lustify for example from civitai?

Felipe Orozco

maybe stupid question, but if i now have 12gb vram and install the appropriate model, if i upgrade to 24gb ram later on would i have to delete and reinstall using the better model?

Robert Johnson

I also checked the win32 folder as someone suggested, and nothing there as well.

Dv

I fixed my issues by following the recommendations outlined for the V2 while installing the V3. It works for now!

Robert Johnson

I have ran the installer and everything seemed to install fine, however, I cannot find the workflow folder or the json file. Maybe I missed something?

Dv

I tried to install this and went through the installation. However, after updating the comfyui and installing the missing nodes, some nodes are STILL missing and they aren't available in the missing nodes window anymore, as if they are installed. When I click on queue for any workflow that has all of the nodes, nothing happens. Nothing starts queing. Any ides?

George Strnad

Seems HUNYUAN isn't loaded, but, I think this one click deal already loaded it automatically no?

George Strnad

Prompt outputs failed validation UnetLoaderGGUF: - Value not in list: unet_name: 'fast-hunyuan-video-t2v-720p-Q8_0.gguf' not in ['flux1-dev-Q5_K_S.gguf', 'flux1-dev-Q8_0.gguf', 'flux1-fill-dev-Q8_0.gguf', 'flux1-schnell-Q5_K_S.gguf', 'flux1-schnell-Q8_0.gguf']

John Cabrer

Can someone guide me to information on how to specify which GPU is used at startup? I have two and it selects the first one, which has less VRAM.

NewBe2

Thats not obvious since most other installers can be launched from downloads.

BadgerWi Johnson

I can't get the manager to show up. Any help would be great thanks in advance. . D:\Flux v3\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build [START] Security scan [DONE] Security scan ## ComfyUI-Manager: installing dependencies. (GitPython) [notice] A new release of pip is available: 25.0 -> 25.0.1 [notice] To update, run: D:\Flux v3\ComfyUI_windows_portable\python_embeded\python.exe -m pip install --upgrade pip ## ComfyUI-Manager: installing dependencies done. ** ComfyUI startup time: 2025-02-14 11:38:16.326 ** Platform: Windows ** Python version: 3.12.8 (tags/v3.12.8:2dc476b, Dec 3 2024, 19:30:04) [MSC v.1942 64 bit (AMD64)] ** Python executable: D:\Flux v3\ComfyUI_windows_portable\python_embeded\python.exe ** ComfyUI Path: D:\Flux v3\ComfyUI_windows_portable\ComfyUI ** ComfyUI Base Folder Path: D:\Flux v3\ComfyUI_windows_portable\ComfyUI ** User directory: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\user ** ComfyUI-Manager config path: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini ** Log path: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\user\comfyui.log WARNING: The script f2py.exe is installed in 'D:\Flux v3\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. [notice] A new release of pip is available: 25.0 -> 25.0.1 [notice] To update, run: D:\Flux v3\ComfyUI_windows_portable\python_embeded\python.exe -m pip install --upgrade pip Prestartup times for custom nodes: 13.4 seconds: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager Checkpoint files will always be loaded safely. Total VRAM 12287 MB, total RAM 32751 MB pytorch version: 2.6.0+cu126 Set vram state to: NORMAL_VRAM Device: cuda:0 NVIDIA GeForce RTX 3060 : cudaMallocAsync Using pytorch attention ComfyUI version: 0.3.14 [Prompt Server] web root: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\web Traceback (most recent call last): File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2113, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait\__init__.py", line 1, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait\nodes.py", line 11, in import dill ModuleNotFoundError: No module named 'dill' Cannot import D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait module for custom nodes: No module named 'dill' Traceback (most recent call last): File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2113, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\__init__.py", line 1, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\nodes.py", line 18, in from .liveportrait.live_portrait_pipeline import LivePortraitPipeline File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\liveportrait\live_portrait_pipeline.py", line 16, in from .utils.filter import smooth File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\liveportrait\utils\filter.py", line 3, in from pykalman import KalmanFilter ModuleNotFoundError: No module named 'pykalman' Cannot import D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ module for custom nodes: No module named 'pykalman' Traceback (most recent call last): File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2113, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\__init__.py", line 8, in import manager_server # noqa: F401 ^^^^^^^^^^^^^^^^^^^^^ File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\glob\manager_server.py", line 13, in import git ModuleNotFoundError: No module named 'git' Cannot import D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager module for custom nodes: No module named 'git' Traceback (most recent call last): File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2113, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node\__init__.py", line 23, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node\nodes.py", line 16, in from segment_anything import sam_model_registry ModuleNotFoundError: No module named 'segment_anything' Cannot import D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node module for custom nodes: No module named 'segment_anything' Import times for custom nodes: 0.0 seconds: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py 0.0 seconds (IMPORT FAILED): D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager 0.0 seconds: D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale 0.0 seconds (IMPORT FAILED): D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ 0.0 seconds (IMPORT FAILED): D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait 1.4 seconds (IMPORT FAILED): D:\Flux v3\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node Starting server To see the GUI go to: http://127.0.0.1:8188

Destin 운명

when I use the installer it says installing git and then it does its thing for a while but closes out shortly after and nothing happens

Travis

This works for me on a Macbook

MarcAndreMueller

i dont know what happened but my comfy ui folder is only 43gb after i ran the installer. its a mistake right? nothing works

Rohit

Did you solved it?

Joseph Beaman

Does anyone know if there are any issues installing the Ultimate Installer on MacOS? Also, how much space does it take on the drive?

Anthony M

Does the V3-MODELS-ULTIMATE_AUTO_INSTALL.bat file need to be a part of this workflow? When I follow the instructions from Exclusive Patreon Video: Troubleshooting Flux Workflow Comfyui! It seems to install ComfyUI_windows_portable again which is redundant.. Perhaps V2 and V3 do not share the same method?

LW

Worked for me but I did have to try a second clean install. Not sure why it failed to install the missing models the first time. Maybe because I didn't just leave the computer to finish. Anyway, the second time I did just let it get on with it (there are lots of models, so it takes a long time). Second time worked. Still had to install missing nodes of course. That took two or three rounds, but again, in the end it worked.

Alex20 Syn

Comment and question Comment: You should mention that the V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL.bat installer dumps the comfyUI folder into whatever folder you run it it. I know that should be obvious but then you'll get idiots like me who just run it from their download folder and spend way to much time thinking the install didn't work. Question: What is V3-MODELS-ULTIMATE_AUTO_INSTALL.bat? Do I need to run this? From where? In the checkpoints folder? Do I just need to download it and have it in the same folder as the installer? What does this bat do?

Timbo hay

https://huggingface.co/Comfy-Org/HunyuanVideo_repackaged/tree/main/split_files/text_encoders

Timbo hay

https://huggingface.co/Kijai/HunyuanVideo_comfy/tree/main

Timbo hay

so all those safetensors files are models, if you look them on google it will take you to a page to download them, the auto downloader dosnt seem to be working, so if you download all the files it says its missing and pop them in the models folder in the relavent areas then it should start working

d

ive literlaly wasted like three nights in a row of this shit like i just have more and more errors. so atp i factory reset my computer so that literally nothing is on ts all i did was the patreon one click installer and it finished and it doesnt fkn work i installed the nodes finally without error so when it starts up it tells me Please 'pip install xformers' which idk if thats normal but it still opens up fine with no other errors and theres no import failed and all the nodes seem to be in place. it also says 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: C:\Users\tcoee\AI\ComfyUI_windows_portable\ComfyUI but i looked it up and it says u can just ignore this? anyway thats all it says when opening it up. then when i try to run anything, like for example i go to do outpainting i test the outpainting bounds and that works then when i actually press queue for the actual outpainting it just doenst work and the console says Failed to validate prompt for output 1310: UNETLoader 1321:0: Value not in list: unet_name: 'flux1-fill-dev_fp8.safetensors' not in ['flux1-dev-fp8.safetensors', 'hunyuan_video_720_cfgdistill_fp8_e4m3fn.safetensors', 'hunyuan_video_FastVideo_720_fp8_e4m3fn.safetensors'] Output will be ignored i mean idr know what to do at this point cuz i have nothing on this laptop other than this damn program!!! so i can keep reinstalling and doing whatever but i swear its js never gonna work i mean. so yea if anyone can help pls

Ron Harris-White II

I am noticing that ADetailer isn't loading. I am getting a redbox around it. I have made sure that it is installed in the Custom Node in my ComfyUI install. Any ideas?

Jeremy A

With so many models of stable diffusion out, I need a starting point. I just upgraded to 4070 ti super 16gb and am starting over. What do you recommend I download first?

Aitrepreneur

just delete the models and files (don't forget to empty the bin to make space on your disk)

Aitrepreneur

you're using the older installer, redownload the new one from the page (same name)

Aitrepreneur

yes I will, you need specific lora training for the hunyuan model, hunyuan won't work with flux loras

Aitrepreneur

https://pastebin.com/NL8Z9QXa I recommend using the installer though, there are a lot of things to install

Timbo hay

same problem it didnt install any models, im trying to get them manually

Bradford Klipan

Hello. I just wanted to thank AItrepreneur for this valuable info! The package and workflow was a bit bloated and took a long time to install considering all I wanted was the Hunyuan A video generator. But the code and info he provided was a very good starting point! I pared down the code and combined it with some other code I found on YouTube and a workflow I found online, and I upgraded the video model to bf16, and now I have a Hunyuan video generator that takes up only 90GB instead of 400GB. But my point is that this is a very good starting point and intro for doing AI video and a variety of other things as well!

Robbie Zek

what are the steps to install this without using the bat file.

ctrl shift face

hey, will you do a tutorial for hunyuan and lora? i tested it, it works no problem with flux but it does absolutely nothing with hunyuan. no matter how simple or complicated the prompt is it absolutely ignores the lora and makes whetever face it wants.

Freelan cedd

Having a hard time installing... with the one bat file installer... it install everything but all the models? a exsample here... It never gets finishe (Attempt 1 of 3 to download clip_l.safetensors... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 3035 100 3035 0 0 298 0 0:00:10 0:00:10 --:--:-- 748 Warning: Problem : HTTP error. Will retry in 10 seconds. 3 retries left. 0 0 0 0 0 0 0 0 --:--:-- 0:00:09 --:--:-- 0)

T

For anyone else running into this- seems like restarting the pod resets some package versions-- re-run just the 'pip install' lines of the auto-install script (someone else mentioned its specifically the torch version that needs fixing but doesn't hurt to update the others too). Afterwards, remember to start the comfyui (using the restart button, not just your browser refresh). I wish Aitrapeneur would just make a new container image for us instead of this using an old base + running install script..

T

Hunyuan video generation (option A) was working yesterday- I stopped the pod when I went to bed, turned it back on today and now every video it produces is just a black screen :(

Jean Dupont

Hello again... I must of done things the perfect wrong way. Nothing works and i have over 1To of différents kinds of models. I was wonderring if there is a "right way" to uninstall the models? Or do i have to delet all the folder and start from scratch?

Jean Dupont

Thanks a lot! It's a first time for me... I see already that i must of done some serious BS! XD i'm over 1To already...

Tom Niestroy

IIRC around 160 for plain Q5 for example to 350 gig depending on your choices. Like with schnell and / or fp8. My Main folder is 359gig ATM.

Dev4life

Thx!

Freelan cedd

Thx. but where do u find that file? sry for the noob style question.

Jean Dupont

Hello to all! Could anyone tell me how much does the full installation weight? I'm in the process and wonderring if i'll have enough space...

pventura

I made a fresh install following latest BAT files and it worked fine except for "LIVEPORTRAIT IMAGE TO VIDEO" and "LIVEPORTRAIT VIDEO TO VIDEO" sections where "ComfyUI-LivePortraitKJ" import failed. I'll take a look later.

pventura

https://www.patreon.com/posts/ultimate-all-in-121355089

Dev4life

Hi, where is V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json file?

JimmyForAI

I am running the GGF Hunyuan Video to Video section of the V3-Ultimate workflow with the Unet model: fast-Hunyuan-video-t2v-720p-Q8_0 (no changes to the other settings except loading the 2-second mp4 video (720x720) I loaded and get this error: - SamplerCustomAdvanced Given groups=1, weight of size [3072, 16, 1, 2, 2], expected input[17, 4, 1, 50, 90] to have 16 channels, but got 4 channels instead. I do not know where I went wrong. I tested changing the VAE_name in the Hunyav2vparam to -taesd3, taef1 and got this message instead: VAEDecodeTiled Expected 3D (unbatched) or 4D (batched) input to conv2d, but got input of size: [1, 16, 1, 16, 16] As far as I can tell, taesd and taesdxl settings lead to the SamplerCustom Advanced error, and the Taesd3 and taef1 lead to the VAEDecodeTiled error.

T

When I try to run hunyuan video option B, the videoCombine step failsbecause it is looking for Error opening output file /workspace/ComfyUI/output/Hunyuan/videos/lowres/vid_00003.mp4. instead of the .png that the previous node creates- how do i fix it?

Hu$h

Thanks bro, I'm still learning comfy been a SD Forge guy for awhile. Having this as a workflow has helped me understand more how nodes and there decencies work a lot more. Never knew the conflicting nodes yellow sign was a general message until yesterday. Always though it meant it would conflict with my currently installed nodes so was just stupid careful until I understood that. lol Good job bro, glad you are back. Wishing you peace of mind, and good health bro. Keep your head up.

Aitrepreneur

yup you can use any workflow you want, everything will always work

aprisma

Is this a normal system message? [INFO] Installing vs_BuildTools... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 curl: (35) schannel: next InitializeSecurityContext failed: CRYPT_E_NO_REVOCATION_CHECK (0x80092012) - Die Sperrfunktion konnte keine Sperrprüfung für das Zertifikat durchführen. [ERROR] Download failed. Please restart the installer Drücken Sie eine beliebige Taste . . . Git already installed. Downloading ComfyUI... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 100 1573M 100 1573M 0 0 30.9M 0 0:00:50 0:00:50 --:--:-- 30.1M Extracting ComfyUI...

Vilain le Singe

Weirdly, Inpainting doesn't work. It just don't use the prompt, it just redux my existing picture. Blocking the seed and changing prompt create the same image over and over. If i change the model from fill to dev while keeping the denoise at 1.0, the masked part is redux hard, but still not using one word of the prompt :0 Help me Overlord !

hubba1987

Just reimport the workflow file

arkcle arf

me too, the same problem. solution 1: that I found: move the folder to another hard drive once the installation is finished, solution 2: To solve this problem, you can add these lines at the beginning of your script, just after @echo off : @echo off cd /d "%~dp0" setlocal enabledelayedexpansion The instruction cd /d "%~dp0" forces the script to use the directory where the .bat file is located as the working directory: %~dp0 is a special variable that contains the path to the directory containing the script. /d allows you to change both drive and directory in a single command. Thus, even in administrator mode, the script will run in the folder where it is located rather than in System32. I suggest you add this line at the beginning of the script to ensure that the installation will always be done in the right place, whether the script is launched in administrator mode or not.

zetta ze

Help please, I cannot find an answer to this.

Freelan cedd

Where do you get the Workflows inside the this workflow? I got alle the Nodes installed but it comes up with this error every time I launch it. ( Missing Node Types When loading the graph, the following node types were not found workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV

Jpvn80

Hello. i am having issues with the python packages. right on the facexlib one an error: "ERROR: could not install packages due to an OSError: [Errno 13] Permission denied: (...)" "consider using the --user option or check permissions." can you help please? if i install it as admin to system32 folder no problem. but if i install it on a diferent folder and hardrive i allways get this.

Alexandre Graça

no luck here, still missing nodes workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm

arkcle arf

look at pventura message, download the new one Aitrepreneur's answer “Aitrepreneur Ok guys fixed, I put the other installer back, download the "V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL.bat" file for a first time install and "V3-MODELS-ULTIMATE_AUTO_INSTALL.bat" is when you want to install other models (if you changed your mind on some of the choices before the install). For now, just delete the folder, dl V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL.bat , launch it normally (not as admin) and let it run ;)"

James Stewart

BTW here's a quick copy of paste of the file running on the command prompt. Seems to be making progress but always hard to tell when you're running something (and your brain starts playing tricks on you when running this long!): ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1136 100 1136 0 0 2333 0 --:--:-- --:--:-- --:--:-- 2337 100 49 100 49 0 0 75 0 --:--:-- --:--:-- --:--:-- 75 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1134 100 1134 0 0 1316 0 --:--:-- --:--:-- --:--:-- 1318 100 49 100 49 0 0 46 0 0:00:01 0:00:01 --:--:-- 46 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1134 100 1134 0 0 2351 0 --:--:-- --:--:-- --:--:-- 2357 100 49 100 49 0 0 76 0 --:--:-- --:--:-- --:--:-- 76 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 etc

James Stewart

Actually the jump/reinstall of everything from v1 to v3 is going for literally almost 24 hours now. Running a NVIDIA 3080 Ti 16gig vram (not sure if that matters). Does this sound right or should I be doing something differently?

HealingPaint

Also the LTX-models were not installed. I had to manually install them through the MODELS search in ComfyUI, just a FYI

Alexandre Graça

same here

DeathPenalty

After initial instructions for regular install i am receiving this message. File "importlib\__init__.py", line 90, in import_module SyntaxError: source code string cannot contain null bytes

James Stewart

Thanks again @Aitrepreneur for your amazing content and thanks for hanging in there for your community. Hope your recovery continues. I'm jumping from Ultimate workflow v1 to v3. Taking a while, but hopefully it gets there soon (been going for hours and hours now as I'm reinstalling I think pretty much everything). Btw have you seen this guy's Hunyuan workflow which includes faceswap (https://www.youtube.com/watch?v=XANTZXPS2Ww) - he says he's happy for other's to use it. Would love to see how it could be included into Ultimate v4, ideally with more than 1 faceswap (FYI I never installed the v2 workflow as it only included 3 faces, ideally I'd like to have 5, or the user know how to change it.)

HealingPaint

I was able to fix it (tested the new inpaint module only so far) by dragging in the V2-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW_N20 WORKFLOW, updating those custom missing nodes (twice, the last 1 found 1 more to update), then restarting comfy entirely..why do I have to install V2-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW_N20 first. Shouldn't the 3.0 do all of this stuff? Just a FYI if that can be fixed that would be great...

HealingPaint

Sadly not working for me. I reinstalled it fresh from the installer you mentioned (after deleting my entire comfy folder to start over), ran it and in Manager updated all for comfy, restarted, installed all missing custom nodes (27, but it failed on two), restarted it again, then it updated those two fine, restarted it and nothing generated anywhere. Sigh. When opening it it pops up missing these modules: workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm I thought this process was supposed to install everything. What am I missing here?

anhvuhoang2390

I got some problem with those plugins, they show Import Failed and tell me Try Fix WAS Node Suite comfyui-reactor-node ComfyUI Essentials ComfyUI-Florence2 ComfyUI-LivePortraitKJ ComfyUI-HunyuanVideoWrapper x-flux-comfyui

dattrax

Do not use pytorch2.6 with this as there's some incompatibilities with hunyuanvideowrapper and the triton package. 2.6 requires triton 3.2, which has removed triton.ops. There's an issue in the triton github about incompatibilities with pytorch2.6. Anyway I just reverted to 2.5 and it seems to load now

Nuno Coelho

I've tested everything and it works fine... with the exception of a few things I cant seem to fix. can you help please? -For the HUNYUAN parts I get some sort of error related to what I think is the text encoder: "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' (F:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\timm\data\__init__.py)" - The liveprortrait img to video and video to video gives me "height and width must be > 0" and "Input and output sizes should be greater than 0, but got input (H: 512, W: 512) output (H: 0, W: 0)" and... it is? - Finally, for the PulidFluxInsightFaceLoader gives me an - **Exception Type:** AssertionError Please let me know if you have an idea on how to fix any of these. very appreciated. thank you for everything!

Aitrepreneur

Ok guys fixed, I put the other installer back, download the "V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL.bat" file for a first time install and "V3-MODELS-ULTIMATE_AUTO_INSTALL.bat" is when you want to install other models (if you changed your mind on some of the choices before the install). For now, just delete the folder, dl V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL.bat , launch it normally (not as admin) and let it run ;)

HealingPaint

^^ thx

HealingPaint

See this plz...we are having issues with this... [...] 100 1140 100 1140 0 0 3346 0 --:--:-- --:--:-- --:--:-- 3362 100 49 100 49 0 0 135 0 --:--:-- --:--:-- --:--:-- 135 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1132 100 1132 0 0 7082 0 --:--:-- --:--:-- --:--:-- 7075 100 49 100 49 0 0 242 0 --:--:-- --:--:-- --:--:-- 242 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1132 100 1132 0 0 6825 0 --:--:-- --:--:-- --:--:-- 6860 100 49 100 49 0 0 263 0 --:--:-- --:--:-- --:--:-- 263 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... It just keeps hammering this, attempting and failing...can you fix this plz??

HealingPaint

Same issue...fix please.

Keith Atkinson

Same here.

HealingPaint

Found it, thanks, wiped my old folder entirely and it's all installing again. Hopefully, it works. :-)

Brushido

Json is in the previous post

HealingPaint

UGH! Where the heck is V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json? I can only find the Runpod one, but I need the local version. I installed Comfy clean and it didn't autolaunch, had to install comfy manager manually and update it, and no JSON..where is it?

Thomas Collard

same issue for me, he redownloading the same file over and over again. EDIT: relaunched everything and got the same issue again.

HealingPaint

Don't you mean for "(works better if you got more than 24gb of Vram)" this? (works better if you got LESS than 24gb of Vram)? Schnell is for easier system requirements, right?

Ciaran Henderson

Okay, I'm dumb I was using your one click installer to setup but I will try to see If I can get this running, lol. Thanks!

arkcle arf

Exactly the same thing for me, please help me.

Sleep Brandynette Bambi

Installation complete, ConfyUI was loaded properly with the nodes. So nice! <3 Now i dont know how to use ConfyUI jet ghagaga geee, im so eager to push out BambiSleep images but i have to work noooo

pventura

I have an strange issue, the installer got stuck on: [...] 100 1140 100 1140 0 0 3346 0 --:--:-- --:--:-- --:--:-- 3362 100 49 100 49 0 0 135 0 --:--:-- --:--:-- --:--:-- 135 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1132 100 1132 0 0 7082 0 --:--:-- --:--:-- --:--:-- 7075 100 49 100 49 0 0 242 0 --:--:-- --:--:-- --:--:-- 242 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... ** Resuming transfer from byte position 4893934904 % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1132 100 1132 0 0 6825 0 --:--:-- --:--:-- --:--:-- 6860 100 49 100 49 0 0 263 0 --:--:-- --:--:-- --:--:-- 263 Successfully downloaded t5xxl_fp8_e4m3fn.safetensors Attempt 1 of 3 to download t5xxl_fp8_e4m3fn.safetensors... [...] Edit: I made a fresh install following @Aitrepreneur latest BAT files and worked fine except for "LIVEPORTRAIT IMAGE TO VIDEO" and "LIVEPORTRAIT VIDEO TO VIDEO" sections where "ComfyUI-LivePortraitKJ" import failed. I'll take a look later. P.S.: I forgot to thank @Aitrepreneur. Good work, man.

Tom Niestroy

would it be possible, to make a batch, to install all models at once? It's kinda jarring to have to run it every time for Q4 with fp8, Q5 and Q8. So a all in one would be nice.

Hu$h

I've been using this since it dropped, loving it, and everything is working smooth. I'm wanting to mess around with your SD 3.5 Large workflow for more style focused stuff. Am I good to run the "SD3-5MODELS-INSTALL-V2.bat" without breaking anything? Just would rather ask then find out the hard way since I started with this Flux V3 workflow first. Thanks bro. Much love.

Aitrepreneur

yes but you need to use the special AMD zluda comfyui: https://github.com/patientx/ComfyUI-Zluda

Aitrepreneur

It's possible sure I'm just too lazy to add colors :D

Aitrepreneur

checkout my troubleshooting video if it's your first time: https://www.patreon.com/posts/exclusive-video-116007628

Aitrepreneur

just re drag and drop the workflow inside comfyui again

Aitrepreneur

It's because you ran the installer as admin, delete the folder and try again this time running the installer normally

Jru Weaver

I have a RX 7900XTX. Is there a way to use it?

Sleep Brandynette Bambi

na honestly, some colors would be nice but i dunno if .bat can have colors. im bash kinda gal, ya kno?

Sleep Brandynette Bambi

installer working like a charm taking awhile tough butt that's due to my 300MBps from home. Love the terminal output bro, easy readable? no, exactly, how i liek it! ;)

Jon Smith

It says I'm missing a bunch of workflows but nothing anywhere mentions it. Where are they?

Michael Moeller

That's a direct link but it's right in the Ultimate All-In-One Flux Workflow V3 post near the bottom Show More there it is.

Michael Moeller

https://www.patreon.com/file?h=121355089&m=415789863

Arkadiy Grigoryants

Hey! Just wanted to say a huge thank you! I finally figured everything out, and now everything is working perfectly. Your workflow is amazing, and you're the best! Appreciate all your hard work!🚀💡

toxrednil

Where is the .json file?

Ish Hodaszi

I've installed all missing custom nodes, and now I have nothing in that list, but comfy is still complaining about some missing node types... I am new to comfy... how do I fix this?

DoraLora Channel

me too, and i just sent u dm.

Claudinei

How do I change the location where webui will be installed here it was installed in the location "C:\Windows\System32" I need to change it to D:

Roybot

when i try to do a new install the manager doesn't load after install. I tried multiple ways. Should I uninstall the comfyui that I installed with Pinokio is it somehow interfering?

Chris Christopher

I'm not sure what I did but it is all working now! Cables still look messy though.

jamal

that seems to have worked. i must have skimmed over that part my apologies. amazing workflow. thank you for all your hard work you are the best!

Chris Christopher

Also all the cables look messy, not like yours in the video which is very neat and tidy

Chris Christopher

thank you for your help! I hope you are feeling better. One last question. I did a fresh install and was able to update everything and install the missing nodes but when the browser opens it says there are missing nodes and I can't seem to generate even the sample in the generate image area. install missing custom nodes is empty. I get this error box saying the following node types are missing: workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm

Aitrepreneur

as I said in the post, this V3 worklow needs the brand new comfyui torch version to work. So the install needs to be done from scratch in a new folder. You can move all the models and files you got from one install to the new one

Aitrepreneur

the problem is the image, it's either the format, webp or the image iself is corrupter. Open the image with paint and save it at png and then load it

Aitrepreneur

you can add a black and white photo in the second image for example to merge the style

Aitrepreneur

send me a dm

Aitrepreneur

that is very strange....never heard of it, can you send me some screenshots in dm and more info about the install?

Aitrepreneur

As answered on discord, vastai or runpod is the same thing, since I haven't used Vast Ai I can only recommend runpod but the principle stays the same

Aitrepreneur

that means that there were some issues during install then, do you see any gguf models downloaded in the comfyui/models/unet folder?

Aitrepreneur

that error is ok, it doesn't impact in anyway the use of the worklofw

Aitrepreneur

the reason why the workflow is "already loaded" is because it uses the cookies in your browser to "save" the latest workflow used, hence why even when you uninstalled everything, the last worfklow gets loaded. If you don't want that, just delete the cookies from that local url

jamal

It's giving me so many issues because the bat file keeps trying to install what is already there from the previous version and it just stops downloading. isn't it possible to just get the .json file, drag it to comfyui and then install missing nodes?

VladoPortos

Fixed it self when I run the installer again this evening :)

BadgerWi Johnson

That seems to have done the trick now I just need to wait for all the comfy sty to install. Thanks for the help

BadgerWi Johnson

Let me give it a shot

Walker4k

Image to Vid Gen Issue - got prompt Failed to validate prompt for output 1867: * LoadImage 1771: - Custom validation failed for node: image - Invalid image file: fds.webp Output will be ignored Failed to validate prompt for output 1767: Output will be ignored WARNING: [Errno 2] No such file or directory: 'Q:\\ComfyUI_windows_portable\\ComfyUI\\input\\fds.webp' Using pytorch attention in VAE Using pytorch attention in VAE VAE load device: cuda:0, offload device: cpu, dtype: torch.bfloat16 Requested to load AutoencoderKL loaded completely 2071.0999996185305 470.1210079193115 True CLIP/text encoder model load device: cuda:0, offload device: cpu, current: cpu, dtype: torch.float16 clip missing: ['text_projection.weight'] !!! Exception during processing !!! GGUF magic invalid Traceback (most recent call last): File "Q:\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 "Q:\ComfyUI_windows_portable\ComfyUI\execution.py", line 202, in get_output_data return_values = _map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "Q:\ComfyUI_windows_portable\ComfyUI\execution.py", line 174, in _map_node_over_list process_inputs(input_dict, i) File "Q:\ComfyUI_windows_portable\ComfyUI\execution.py", line 163, in process_inputs results.append(getattr(obj, func)(**inputs)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "Q:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF\nodes.py", line 152, in load_unet sd = gguf_sd_loader(unet_path) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "Q:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF\loader.py", line 25, in gguf_sd_loader reader = gguf.GGUFReader(path) ^^^^^^^^^^^^^^^^^^^^^ File "Q:\ComfyUI_windows_portable\python_embeded\Lib\site-packages\gguf\gguf_reader.py", line 95, in __init__ raise ValueError('GGUF magic invalid') ValueError: GGUF magic invalid Prompt executed in 1147.68 seconds --------------------------------- AI tells me - This error log actually contains two distinct issues that cause processing to fail: The image loading part fails because the workflow attempts to load an image file named “fds.webp” from the directory Q:\ComfyUI_windows_portable\ComfyUI\input but the file is either missing or invalid. The log shows a validation error at the “LoadImage” node, with the message “Invalid image file: fds.webp”, followed by a Python warning indicating that the file cannot be found (“[Errno 2] No such file or directory”). In effect, the prompt that depends on this image is disregarded because the system cannot locate or properly read it. Later in the log, the error “GGUF magic invalid” is raised during the attempt to load a model using the GGUF loader. When the function attempts to create a GGUFReader instance for the unet model file, it checks for a valid “magic number” (a specific signature in the file header that identifies a proper GGUF file). In this case, the file did not match the expected signature and so the loader raises a ValueError. This “magic” check is a standard way to ensure that the file is of the correct format and not corrupted or mistakenly replaced by another file type. In summary, the workflow fails for two reasons: first, because the input image “fds.webp” is missing or corrupt, and second, because the unet model file supplied to the GGUF loader isn’t recognized as a valid GGUF file (its magic number doesn’t match). Both issues cause the system to abort processing of the prompt.

Michael Moeller

This error I have also encountered well before using this workflow it appears to have been an error generated from ComfyUI Universal Styler my investigation found that there are three ways to handle this error. You can disable the node, the error messages at loading may still appear, or uninstall it. I've disabled it myself. There is also a third fix by going to this page for ComfyUI Universal Styler https://github.com/KoreTeknology/ComfyUI-Universal-Styler?tab=readme-ov-file, best of luck to you.

Michael Moeller

Have you tried to drag and drop the JSON workflow file into the blank workspace canvas?

Sir Lansealot

Update. I managed to get it working. I had to download the .safetensors file offline, in my case for the Image to Video I needed "ltx-video-2b-v0.9.safetensors". Once I downloaded it and put it into the "Checkpoint", "Unet" and "Vae" folders (in the "Models" folder) it worked. Effectively confirming what we thought, that there are indeed some missing .safetensors files from the install.

ПРОСВЕТЕН

Maybe my AVG is stopping installation, detecting idp.alexa.54

Mia Strong

Hi! How can I silize a photo with Redux, as we can see in the preview "Image Variation and Restyling with FLUX.1 Redux" 8:34 - your last video on Youtube. If I put only one photo and make promt: b&w portrait illustration, I get another PHOTO img. If I use one photo + img with the style I need, I lose the main details of the photo. In the preview we can see how the photo is stylized... how to work with Redux correctly?

Geblendert

That works! Thank you!

Chris Christopher

I tried deleting the folder and reinstalling. Strange thou because once it opened the v3 workflow was already loaded. I feel something isn't getting deleted, like a preference folder somewhere. And then all the errors I got while trying to install the missing nodes got worse and didn't install at all

aprisma

First time that I have some problems with totally new install. Tried it 3 times: ComfyUI version: 0.3.13 [Prompt Server] web root: C:\AI\ComfyUI_windows_portable\ComfyUI\web Traceback (most recent call last): File "C:\AI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2110, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait\__init__.py", line 1, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait\nodes.py", line 11, in import dill ModuleNotFoundError: No module named 'dill' Cannot import C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait module for custom nodes: No module named 'dill' Traceback (most recent call last): File "C:\AI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2110, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\__init__.py", line 1, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\nodes.py", line 18, in from .liveportrait.live_portrait_pipeline import LivePortraitPipeline File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\liveportrait\live_portrait_pipeline.py", line 16, in from .utils.filter import smooth File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ\liveportrait\utils\filter.py", line 3, in from pykalman import KalmanFilter ModuleNotFoundError: No module named 'pykalman' Cannot import C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ module for custom nodes: No module named 'pykalman' Traceback (most recent call last): File "C:\AI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2110, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\__init__.py", line 8, in import manager_server # noqa: F401 ^^^^^^^^^^^^^^^^^^^^^ File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\glob\manager_server.py", line 13, in import git ModuleNotFoundError: No module named 'git' Cannot import C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager module for custom nodes: No module named 'git' Traceback (most recent call last): File "C:\AI\ComfyUI_windows_portable\ComfyUI\nodes.py", line 2110, in load_custom_node module_spec.loader.exec_module(module) File "", line 999, in exec_module File "", line 488, in _call_with_frames_removed File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node\__init__.py", line 23, in from .nodes import NODE_CLASS_MAPPINGS, NODE_DISPLAY_NAME_MAPPINGS File "C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node\nodes.py", line 16, in from segment_anything import sam_model_registry ModuleNotFoundError: No module named 'segment_anything' Cannot import C:\AI\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-reactor-node module for custom nodes: No module named 'segment_anything'

VladoPortos

Sorry to bother you with this, but when I try to run the gpu bat. I get : "Torch not compiled with CUDA enabled" ``` G:\AI Video\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build [START] Security scan [DONE] Security scan ## ComfyUI-Manager: installing dependencies done. ** ComfyUI startup time: 2025-02-06 12:18:05.805 ** Platform: Windows ** Python version: 3.12.8 (tags/v3.12.8:2dc476b, Dec 3 2024, 19:30:04) [MSC v.1942 64 bit (AMD64)] ** Python executable: G:\AI Video\ComfyUI_windows_portable\python_embeded\python.exe ** ComfyUI Path: G:\AI Video\ComfyUI_windows_portable\ComfyUI ** ComfyUI Base Folder Path: G:\AI Video\ComfyUI_windows_portable\ComfyUI ** User directory: G:\AI Video\ComfyUI_windows_portable\ComfyUI\user ** ComfyUI-Manager config path: G:\AI Video\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini ** Log path: G:\AI Video\ComfyUI_windows_portable\ComfyUI\user\comfyui.log Prestartup times for custom nodes: 2.3 seconds: G:\AI Video\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager Checkpoint files will always be loaded safely. Traceback (most recent call last): File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\main.py", line 136, in import execution File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\execution.py", line 13, in import nodes File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\nodes.py", line 22, in import comfy.diffusers_load File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\comfy\diffusers_load.py", line 3, in import comfy.sd File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\comfy\sd.py", line 6, in from comfy import model_management File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\comfy\model_management.py", line 166, in total_vram = get_total_memory(get_torch_device()) / (1024 * 1024) ^^^^^^^^^^^^^^^^^^ File "G:\AI Video\ComfyUI_windows_portable\ComfyUI\comfy\model_management.py", line 129, in get_torch_device return torch.device(torch.cuda.current_device()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "G:\AI Video\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py", line 971, in current_device _lazy_init() File "G:\AI Video\ComfyUI_windows_portable\python_embeded\Lib\site-packages\torch\cuda\__init__.py", line 310, in _lazy_init raise AssertionError("Torch not compiled with CUDA enabled") AssertionError: Torch not compiled with CUDA enabled G:\AI Video\ComfyUI_windows_portable>pause ```

Arkadiy Grigoryants

Hi! Thanks for your hard work on this workflow! I was finally able to load the workflow, but I noticed a critical issue: the enable/disable switches (toggle buttons) for each block are missing. In your video, I can see that you can enable and disable specific parts of the workflow, like the image generation block or the upscaler. But in my setup, those toggle buttons are not visible, which makes it impossible to work efficiently. Could you please guide me on how to restore them? Is this a missing node issue or a setting I need to enable? Thank you in advance!

BenignOne

Hi! Thank you for the amazing help. This pushed the install a tiny bit further but It gets stuck in a loop now... It tries to install visual studio (when i already have it installed) and then 7z (also installed). Ive tried letting it install anyway and trying again but that doesnt work. Ive tried telling it to ignore that process and it doesnt work. Both of these options just end the install process. Any ideas? never had this sort of issue before.. EDIT: SOLVED-I uninstalled the programs, let it install again when suggested, second time around I told it to ignore installs and it all started working. Hopefully this helps others

zetta ze

I installed with option 1, I have an Nvidia 3080 with 10GB. Then option B, no Flux schnell. And B, No FP8.

Arkadiy Grigoryants

HI! I managed to install everything, and now the nodes are loading. However, I noticed that the enable/disable buttons for node groups are missing. This makes it difficult to activate or deactivate parts of the workflow. Is this a known issue, or do I need to change some settings?

toxrednil

I about to take the leap and use Vast.AI to do this because I want to replace Kling image to video creation. Can I use the ComfyUI template on Vast.AI to run this? Or should I use RunPod? Which is better? Waiting for answers before I jump. Please answer.

BadgerWi Johnson

After install it doesn't auto run. And when I open it manually I just get the standard blank comfy UI and not your work flow. Any suggestions?

Ernest Cline

On the GGUF side, 'Unet Loader' I am not able to select from any model options. All I have is 'undefined'. I can select different models just below that in the DualCLIPLoader.

Hometrees

solved in the thread. just exit out the workflow and load it in fresh like maciej said x

Hometrees

same issue workflowFlorence workflowFlux Controlnet workflowXLabs Sampler workflowoutpaintflor workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm

Killy_Blame

[] [] Exception in callback _ProactorBasePipeTransport._call_connection_lost(None) handle: Traceback (most recent call last): File "asyncio\events.py", line 88, in _run File "asyncio\proactor_events.py", line 165, in _call_connection_lost ConnectionResetError: [WinError 10054] Eine vorhandene Verbindung wurde vom Remotehost geschlossen Exception in callback _ProactorBasePipeTransport._call_connection_lost(None) handle: Traceback (most recent call last): File "asyncio\events.py", line 88, in _run File "asyncio\proactor_events.py", line 165, in _call_connection_lost ConnectionResetError: [WinError 10054] Eine vorhandene Verbindung wurde vom Remotehost geschlossen FETCH DATA from: C:\Ai\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager\extension-node-map.json [DONE] 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: C:\Ai\ComfyUI_windows_portable\ComfyUI 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: C:\Ai\ComfyUI_windows_portable\ComfyUI 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: C:\Ai\ComfyUI_windows_portable\ComfyUI 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: C:\Ai\ComfyUI_windows_portable\ComfyUI 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: C:\Ai\ComfyUI_windows_portable\ComfyUI 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: C:\Ai\ComfyUI_windows_portable\ComfyUI [] []

Aitrepreneur

I could yes, but then you'll have no choice over all the different models. Also if you have to dl a single 200gb archive, you'd better have a good connection because if it fails you can start from scratch

Aitrepreneur

have you chosen to dl the FP8 models during install? If not you need to use the gguf side (on the right)

Aitrepreneur

I'm not sure I understand the issue, what errors? Cleaning it is really easy, just indeed delete the comfyui portable folder and you're done, you can relaunch the installer in a new folder then

Aitrepreneur

reload the workflow, check my troubleshooting video if it's your first time: https://www.patreon.com/posts/exclusive-video-116007628

Aitrepreneur

what options did you chose during the install? In the gguf side all you need to do is just select the models that were downloaded for your GPU setup, no need to add anything

Aitrepreneur

No but in that case you need to install the AMD comfyui version: https://github.com/patientx/ComfyUI-Zluda

Aitrepreneur

https://www.patreon.com/posts/ultimate-all-in-121355089

Aitrepreneur

there must have been an issue during the install and some files might be missing. Have you tried relaunching the installer in a new folder? where there any errors?

Aitrepreneur

all the links are in the post above. here's the link for the V3 workflow: https://www.patreon.com/posts/ultimate-all-in-121355089

Aitrepreneur

So this error can happen when first launching comfyui, all you need to do is just update it, you can go to the update folder and click on the update_comfyui_and_python_dependencies.bat file and update everything, then you can just relaunch it

Aitrepreneur

Sure I don't see why not, I don't do it because it costs more money than simply redoing the install whenever you need it though so it's up to you

Pete

Thank you for your work. Wouldn't it be easier to provide a ZIP file of a working ComfyUI directory with a BAT file inside to install models and dependencies? That way, it's less likely for people to encounter errors.

Michael

I get an a error when trying the new inpaint model Prompt outputs failed validation UNETLoader: - Value not in list: unet_name: 'flux1-fill-dev_fp8.safetensors' not in ['flux1-dev-fp8.safetensors', 'hunyuan_video_720_cfgdistill_fp8_e4m3fn.safetensors', 'hunyuan_video_FastVideo_720_fp8_e4m3fn.safetensors']

Maciej Frymorgen

Those are mostly group nodes. Didnt load for me properly at first as well. Removing workflow and reloading from scratch after all updates fixed it.

Keith Atkinson

I went through the same process, i.e. AUto install, delete and then manual install and get exactly the same missing nodes. I'm sure there will be an simple explanation.

Chris Christopher

I am having really bad issues now. I tried deleting the comfyui portable folder and then running the installer again for "fresh install" but now I get errors everywhere. Wehn it does boot into comfyui it still shows the manager, which I installed manually, and the workflow. Whats strange to me is that I have not yet installed those into the fresh install so something is stored elsewhere on the computer and I don't know how to uninstall everything to be truly clean.

Sir Lansealot

Yeah, that sounds like what I have. I've got a few .gguf files that are missing their corresponding .safetensors files. Hopefully he can help us when he can :)

Arkadiy Grigoryants

Hi! Thanks for your amazing workflow. I'm facing an issue with installation—when loading the graph, ComfyUI gives a "Missing Node Types" error and cannot find: workflowFlorence workflowFlux ControlNet workflowNodes 3 UltimateSDUpscale workflowXLabs Sampler I tried installing them manually via git clone and also used the "Install Missing Custom Nodes" option in the manager, but the nodes still don’t appear. I'm using the latest version of ComfyUI on Windows 10 with an RTX 4070 Ti. Could you please help? Maybe the folder paths or node names need to be different?

Kristopher Dean

A lot of the custom nodes are still missing after the update... help?

zetta ze

Installed clean from scratch everything seems to have installed correctly. I did have to manually copy v1-5-pruned-emaonly-fp16.safetensors file in the checkpoints folder. But trying a simple image generation in GGUF it goes up to about 75% done and then gives me CLIPTextEncode 'NoneType' object has no attribute device. I tried replacing the DualCLIPLoader(GGUF) with TripleCLIPLoader but I can't get it to run. What am I missing, and what tensor files am I suppose to use in the DualCLIPLoader(GGUF)?

rubz

title: ULTIMATE ALL-IN-ONE FLUX WORKFLOW V3!

Ciaran Henderson

I got this error: "RuntimeError: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx" I have a 7900XTX, does this app only work with Nvidia GPU's? :(

Kapral Patrick Sepp oma abikaasaga

yep got most of them working by reinstalling but still have some safetensors missing. idk why the installer dosnt download them. maybe @Aitrepreneur knows fix to it or .bat to redownload them :)

Sir Lansealot

It should be on the "Recent Posts" segment, directly under this post. Click on the "Home" tab on his page and scroll down a bit.

Tiago Fernandes

Second edit: Everything seems to be working great. I had a problem with SDUpscale node but I uninstalled it and reinstalled it manually and now it's fine. Thank you for the workflow, @Aitrepreneur. ================================================================================== Edit: I uninstalled 7zip because it was in another drive (I like to keep my pc organized, but whatever...) and changed to default (meh) directory and the installation is working. I still got the problem with Schnnel. When the thing is over I'll test everything and I'll see if it is working fine and I come back here to edit this message once more. ================================================================================== I can't even start the installation. For some reason, it does not detect the 7zip in my PC, even though I do have it installed. And before that I also got this message: "curl: (35) schannel: next InitializeSecurityContext failed: CRYPT_E_NO_REVOCATION_CHECK (0x80092012) - The revocation function was unable to check revocation for the certificate. [ERROR] Download failed. Please restart the installer" It does not matter if I decide to install or not Schnnel; I guess I'll have to install it manually and just use the workflow.json...

Brad Tayan

I'm scrolling and checking older posts, and I still can't find V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json

Brad Tayan

I can't find it either.

EW

I'm also getting this.

ПРОСВЕТЕН

Hello. Thanks for your work!!! I have an issue - I installed all with auto installation but I didn't have Manager. I installed it with instructions, and with manager added all other files, when I restarted all I Cant start again with bat file and received this message: D:\AI\ComfyUI_windows_portable>.\python_embeded\python.exe -s ComfyUI\main.py --windows-standalone-build Traceback (most recent call last): File "D:\AI\ComfyUI_windows_portable\ComfyUI\main.py", line 1, in import comfy.options ModuleNotFoundError: No module named 'comfy.options' D:\AI\ComfyUI_windows_portable>pause Press any key to continue . . . Please help - what to do?

Bruce

So mistakes were made on my end. And I no longer have the workflow. Is there a download for it?

Bradford Klipan

Can this be installed on a persistent network volume on Runpod?

Jérémie Aucher

Essaye ça: 1) Install-PackageProvider -Name NuGet -Force | Out-Null 2) Install-Module -Name Microsoft.WinGet.Client -Force -Repository PSGallery | Out-Null 3) Repair-WinGetPackageManager -Force -Latest

sealsnipe

same here

Chris Christopher

I deleted everything and am reinstalling and am getting warnings. " WARNING: The scripts sz_split.exe and sz_wc.exe are installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The scripts lsm2bin.exe, tiff2fsspec.exe, tiffcomment.exe and tifffile.exe are installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The scripts imageio_download_bin.exe and imageio_remove_bin.exe are installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The scripts cygdb.exe, cython.exe and cythonize.exe are installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The scripts backend-test-tools.exe, check-model.exe and check-node.exe are installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The script humanfriendly.exe is installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The script coloredlogs.exe is installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The script onnxruntime_test.exe is installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The script onnxruntime_test.exe is installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location. WARNING: The script insightface-cli.exe is installed in 'C:\Users\Cosmo\Desktop\ComfyUI_windows_portable\python_embeded\Scripts' which is not on PATH. Consider adding this directory to PATH or, if you prefer to suppress this warning, use --no-warn-script-location.

Chris Christopher

When I update all I get the message "[ComfyUI Manager] Failed to switch to the default branch ComfyUI update fail: The installed ComfyUI does not have a Git repository." Not sure why. I also did not see any errors on the install and the manager wasn't there so I loaded it manually

Nuno Coelho

Good job! I have a few errors here and there, and would love if we had a 1 click mode install cus I have a lot of them missing :) thanks and keep it up!

Jokerlorian

I found out what my issue was, I was expecting it to be the \python_embedded folder that was the issue not C:\Users\%UserName%\AppData\Roaming\Python It's proceeding now.

Aitrepreneur

it's a known comfyui bug, even when everything is downloaded correctly, you need to re load the workflow to display everything correctly. I've made a special troubleshooting video that I put the link in the post but here's the direct link to it: https://www.patreon.com/posts/exclusive-video-116007628

Aitrepreneur

Where there any issues or error during install? The manager gets installed automatically as will everything else. To get the manager go to the ComfyUI_windows_portable\ComfyUI\custom_nodes folder click on the folder path then type cmd press enter and in the command prompt window that just appeared you're gonna type: git clone https://github.com/ltdrdata/ComfyUI-Manager.git then press enter now if you reopen comfyui the manager will be here

Aitrepreneur

It works, what's the issue you're having?

Aitrepreneur

Are you sure that you are running the installer in a completely empty folder? This error means that it is finding an already existing insightface folder already present.

Jokerlorian

Having problems with InsightFace. So far I've nuked my entire ComfyUI folder and every time the V3.bat runs it has the following error. Collecting opencv-python-headless Using cached opencv_python_headless-4.11.0.86-cp37-abi3-win_amd64.whl.metadata (20 kB) Requirement already satisfied: numpy>=1.21.2 in e:\comfyui\comfyui_windows_portable\python_embeded\lib\site-packages (from opencv-python-headless) (1.26.4) Using cached opencv_python_headless-4.11.0.86-cp37-abi3-win_amd64.whl (39.4 MB) Installing collected packages: opencv-python-headless Successfully installed opencv-python-headless-4.11.0.86 Setting up PulID model... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1144 100 1144 0 0 6520 0 --:--:-- --:--:-- --:--:-- 6612 100 1089M 100 1089M 0 0 39.9M 0 0:00:27 0:00:27 --:--:-- 40.5M Cloning InsightFace repository... Updated Git hooks. Git LFS initialized. fatal: destination path 'insightface' already exists and is not an empty directory. Failed to clone InsightFace repository.

BenignOne

Hi! Thank you for the help, I'm on windows 10, i tried what you said but I got the below. The term 'Repair-WinGetPackageManager' is not recognized as the name of a cmdlet,

Jokerlorian

I had the same problem (Win11) open a Powershell prompt as admin and run "Repair-WinGetPackageManager" (no quotes). That resolved the issue for me. Hopefully that helps your issue as well.

Walker4k

From another user here, it worked for me but be aware, some of these things take ages to download esp after you hit the generate button. A few times I thought the whole thing had stalled :-/ just needed to be more patient - · 2d Quote Timothy X down below "EDIT: Just drag the workflow file into Comfyui and that fixes it." Same issue but on Windows. Fresh install, installed missing nodes, still says I'm missing all workflow type nodes.

Bradford Klipan

Actually I wanted to say that all this crap you gave us is worthless and doesn't even work! Back to square one! Figure out how to install ComfyUI on runpod the correct way!!!!

Pete

is anyone else still not got the manger button after following fixes?

Filip Šenjug

You can edit .bat file and remove the lines where it is downloading the models.

HOD - 3D Model Studio

same here

Vox Efx

When loading the graph, the following node types were not found workflowFlux Controlnet workflowXLabs Sampler workflowpulidtxt workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV

BenignOne

Feeling really stupid here as I usually have no issue installing these things. As soon as I run the .bat file, I get a popup 'Cannot open app package' - 'Reason' - 'The package is either corrupted or invalid.' I went to do the manual install but its not letting me connect to the page / its down?

Sir Lansealot

I've just tried this, but to no success. One thing I forgot to mention was, when I click on the "Load Checkpoint" model, it switches to "Undefined" and I can't change it back.

Walker4k

Any advice on this -**'Int-🔬 (In group node 'workflow>outpaintsize1') workflow>outpaintsize**

Gio

Im also having the bug that the manager button or any others are not showing up in that part of the menu so cant update missing files.

Virtamouse

You sir. Are a treasure. Patiently waiting for your video on how to train hunyuan loras. Cant wait.

MisterDeepvision

LTX model does not load in checkpoint...gives error Failed to validate prompt for output 1711: * CheckpointLoaderSimple 1693: - Value not in list: ckpt_name: 'ltx-video-2b-v0.9-fp8_e4m3fn.safetensors' not in [] Output will be ignored invalid prompt: {'type': 'prompt_outputs_failed_validation', 'message': 'Prompt outputs failed validation', 'details': '', 'extra_info': {}}

Nv Shadow

yo is there a 1 click v3 install model's file

Moses Son

scroll down his post's it should be the one before this one.

Moses Son

scroll down his post's it should be the one before this one.

Nathan Quintanal

Be sure to not only restart ConfyUI and the browser, but to also reload the workflow as the error get stored in cache for some reason. redrag the json to the browser.

Arkadiy Grigoryants

Hello! How to install all the necessary modules from this file MODELS-ULTIMATE_FLUX_INSTALL , in which folder should it be launched? comfyui already installed

O'Neill Joseph

Where do I find the V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json file separately

Arkadiy Grigoryants

I also installed it manually and the manager started working. But another problem arose: how to supply all the necessary modules. In which package should I activate this file? MODELS-ULTIMATE_FLUX_INSTALL.bat

hemanth kumar

Hi. can you please tell me where can I findV3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json file is

Nuno Coelho

Anyone else getting the "Missing Node: EmptyHunyuanLatentVideo" error? even using ther 1 click installer, installing nodes AND updating comfy

Esteban

Thank you soo much for the hard work on the workflow and the scripts, it's really appreciated!

Reign2294

Am I missing something or does the GGUF Depth ControlNet not generate a img-to-img once the depth controlnet mask is created? Pls help...

Que Boyer

Welcome back. hope your health continues to improve!

LW

Just a heads up: You can check out the video right here: https://youtu.be/zLGthxUe2jc Is incorrect; that loads the V2 video... which is definitely worth watching. The correct link to the V3 video is given a bit further down, but for lazy asses like me :) ..... You can check out the video right here: https://youtu.be/q5kpr84uyzc

LW

Just below....

Jonathan

I ACTUALLY got the right fix: I needed to purge the installation cache if I understood it right. You have to do this command: python.exe -m pip cache purge And then it works somehow

Draco The creative

https://www.patreon.com/posts/ultimate-all-in-121355089

George Strnad

where is the .json file?

Mateus Henrique de Sousa Mello

Linux sh install should be good

Jonathan

How do you delete everything? Is it only the ComfyUI portable folder or is it more?

Janne Kallio

I think there is some bug in the liveportrait manual when saving the still image. the video looks ok, but the frame saved for the "preview image" is wrong. it is a bit strange as some images seem to have slight changes, but most are just the original image without any changes. I have to manually save the image to get the changes.

Kapral Patrick Sepp oma abikaasaga

got it to work, so basically delete comfyUI root folder and reistall it again with all the steps.

Kapral Patrick Sepp oma abikaasaga

+1

Jonathan

So I fixed an old problem and now have a new one: When I click on "Update all", I get this error: "Failed to update ComfyUI or several extensions. See terminal log." The error shown in cmd is: PermissionError: [Errno 13] Permission denied. Anyone having a tip?

Jonathan

So I "fixed" this error by reinstalling it as an administrator BUT!! I don't know if this is a godd idea since it's still not working: SOmehow it's not doing anything while I click on "Queue". When I click on "Update all" I get this error: "Failed to update ComfyUI or several extensions. See terminal log." Anyone having a tip?

Raphael Paz

same

E

Clean install solved a lot of my issues. Deleted everything reinstalled now i have the following error for nodes not found workflowfillparam workflowREDUXOTHERPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowLIVEPORTRAITIMGPARAM

Alex Mahr

Where can I find "V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json"

Jovan Kastrol

same with me after comfyui finished downloading the cmd is closed

Timothy X

I'm having the same problem.

Phantomartdrop

Thanks so much for putting this together. Amazing work!

Gamearc

Can you please also include a batch file to download just the models if i alreayd have comfyui installed tnx

goedkoop store

where is the json file ? ?? ? ? ??

CASA CASA

Hi! Can I donwload the workflow alone?

Jorge

I am also getting the same error

M.K

Hello, is there any command i can add to skip downloaded files, cause .bat crashed midway . also i have 6gb vram so im not sure what to download and what would be useless for me .

kristopher ulley

I found a fix download a new workflow file and use it. worked for me

Catalin Fuca

reloading the workflow fixed it for me as well

kristopher ulley

whats the name of the folder?

kristopher ulley

same

Mograine

Instalation keeps crashing.

Javier Lopez

this is what I have now: pytorch version: 2.5.1+cu121 xformers version: 0.0.28.post3

Javier Lopez

Upgrade to pytorch 2.5.1 That solved the issue for me

fri3d

Yep. I just get a blank video file too.

John H

Put this on a fresh install of windows 11. Came out to 230gb woah!

Danny Richard

Getting "Cannot Open App Package" right away says corrupted or invalid.

James Bowler

The install seems so far to be working for me but I did get his error: ####### LivePortraitCropper (In group node 'workflow>LIVEPORTRAITIMGPARAM') workflow>LIVEPORTRAITIMGPARAM LivePortraitProcess (In group node 'workflow>LIVEPORTRAITIMGPARAM') LivePortraitComposite (In group node 'workflow>LIVEPORTRAITIMGPARAM') workflow>LIVEPORTRAITVID2VIDPARAM workflowLIVEPORTRAITIMGPARAM LivePortraitLoadCropper DownloadAndLoadLivePortraitModels LivePortraitLoadMediaPipeCropper workflowLIVEPORTRAITVID2VIDPARAM Because when I check missing nodes, I see this: IMPORT FAILED ↗ ComfyUI-LivePortraitKJ DO NOT try TRY Fix, it will blow up your install, so I will wait for further guidance, THX!

Sir Lansealot

Sorry to be a pain. I've just installed everything no problem, updated ComfyUI and installed the missing nodes, again, no problem. I tried using "FP8 LTX Image to Video" and got the following message in the Command Prompt: got prompt Failed to validate prompt for output 1867: * CheckpointLoaderSimple 1778: - Value not in list: ckpt_name: 'ltx-video-2b-v0.9-fp8_e4m3fn.safetensors' not in [] Output will be ignored Prompt executed in 0.17 seconds I would like to note the "Load Checkpoint" has a red border around it at this point. Based on this, I assume that it can't find the model as it isn't generating a video. One observation I made from watching your latest Youtube video, is that at 1:05 into the video, you have "V3-MODELS-ULTIMATE_FLUX_INSTALL". Do I need to download the models separately, like in V2, or are they part of the "V3 Manager Auto Install" and already downloaded? Once again, sorry to be a pain, and thanks for any help.

FuzzySilk

How do you install the missing nodes? If I delete the workflow, how do I add these nodes before I import the workflow again?

EarNugget

It worked!

Pete

Hello, I was able to get the button to appear using some software called Stability Matrix. I tried transferring the working files from Stability Matrix into the broken portable, but no luck. So, I will try and use this for now.

KinopioKun

Brand new install of this, I ran the batch file and there are several pieces of ComfyUI missing, once I fixed those, I'm getting tons of conflicts when I install the missing custom nodes. No amount of reinstall or different versions fixes it. Is this actually tested? lol

Pete

I installed this in the past and had the manger button not come up and fixed it by using cmd in the nodes folder as normal and using installing manger. However this time everything I try the manger button will not show up. I even asked on git hub and they said to update and still its not showing up. Any chance you can send me a zip and see if that works? Im running out of ideas to get this working :C

ProAlexp

Getting "Can't open AppPaket" at the direct start of installing with the batfile and appinstaller says that the archive is damaged or invalid

Lone Striker

Thank you! This was the last bit that was killing my ComfyUI installation!

Janne Kallio

yep.. download failed somehow at 115gb .. after 10h.. so start from beginning I guess.. :(

Jonathan

Hey guys :), I have a "permission denied" error when trying to run the bat file, it just stops after a few seconds, giving out the error "could not install packages due to an OSError [Errno 13] Permission Denied." Does anybody know what could fix this?

Kevin A

I installed in on runpod and now the manager button disappeared. How to fix that?

Kapral Patrick Sepp oma abikaasaga

if someone knows how to reinstall models again or .safetensors leave a comment! :)

Kapral Patrick Sepp oma abikaasaga

same, downloaded it m,anually

Kapral Patrick Sepp oma abikaasaga

same it like somehow didnt install .safetensors

ChudJones

Thanks for creating this and thanks for adding the simple image upscaler. I did run into a missing checkpoint for the LTX i2v workflow (ltx-video-2b-v0.9-fp8_e4m3fn.safetensors), so I just dl'd it manually.

LeoSolis

Solved mine, after updating everything you need to re-drop json file to comfy that fixes modules. This video of his has many troubleshoot about the version https://www.patreon.com/posts/exclusive-video-116007628

Catalin Fuca

Got the same issue

Hpz937

it would be nice to get the bat file version for linux as well. I don't run windows.

Hpz937

I had this happen too. What fixed it was deleting the workflow and re-importing it after installing missing nodes.

Javier Lopez

Seems like Hunyuan option A requires pytorch 2.5.1

RestlessZombi

https://www.patreon.com/posts/ultimate-all-in-121355089 - It's not so obvious it's in the other post.

Oscar Wojciechowski-Prill

I can't find the V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json file to load. It didn't install with the batch file, I don't see one to download. Am I missing something obvious?

Kerfelox

Failed to validate prompt for output 1310: * UNETLoader 1321:0: - Value not in list: unet_name: 'flux1-fill-dev_fp8.safetensors' not in ['flux1-dev-fp8.safetensors', 'hunyuan_video_720_cfgdistill_fp8_e4m3fn.safetensors', 'hunyuan_video_FastVideo_720_fp8_e4m3fn.safetensors'] Output will be ignored Prompt executed in 0.84 seconds" This is what i get when trying out the inpaint and outpaint feature, how do i solve this?

Mark P

Thank you! Pretty cool workflow! I like that you incorporated the old with the new. Everything is in one place.

Sam Shennan

Can this run on a linux server? I'd like to setup a comfyui for my team at work

Sylvain

I just downloaded the Workflow, had everything already installed. Works pretty smooth. Thanks for your work!

Janne Kallio

yep, with bad internet connection this becomes important.. not saying I am only half way after 6h,,, but I am,,.

Walker4k

Similar to others - Missing Node Types When loading the graph, the following node types were not found workflowfillparam workflowREDUXOTHERPARAM workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV

Janne Kallio

this was good info to know! with lot of work I managed to clear up that amount of space.. but my internet is so slow.. only half way downloading in 6 hours :(

Sonny Driscoll

https://www.patreon.com/posts/ultimate-all-in-121355089

Aaron

Where is the V3-ULTIMATE_FLUX_ALL-IN-ONE-WORKFLOW.json file??

Charles Kwok

If you run Windows and have trouble with InsightFace download, change the CURL line to this: powershell -Command "Invoke-WebRequest -Uri '!WHL_URL!' -OutFile '!WHL_FILE!'"

Arkadiy Grigoryants

Hello! I installed everything according to the instructions, but there is no manager in the menu to update everything

Hans-Joachim Nolte

hello! i need to install it to drive E. is this possible if i change set "winget_path=%userprofile%\AppData\Local\Microsoft\WindowsApps" to a path on drive E?

Yukon Cornelius

I was wondering if in the future you might be able to make a version that doesn't download all the models over again? Like it is downloading llama 8b again, which i already have. and its like unskippable. and i assume its gonna do that for everything. great stuff and i hope it works well. Just wondering for the future.

Mateus Paixão

In the cmd say

LeoSolis

Yep i have the same issue same missing nodes

MarcAndreMueller

how did you find out? thats bad, aren't those big files?

gregoryallen caldwell

will this work on AMD Ryzen 7 5700U with Radeon integrated Graphics, 1801 Mhz 8 core/ 16 processors 32 gigs ram ? if not is there a workaround?

LeoSolis

Even tho i did Update All and install missing custom nodes in UI manager there are still lots of missing custom nodes list error at the start but if i do check update or check missing in manager it's not showing any results. just confused.

Mateus Paixão

for some misterious reason, the bat is droping a folder in system32, with mananger, models, and many things.... idk why

Harly Hallikas

Beware that the installatoin will require at least 233GB of free disk space. I had no idea and tried installing on my smaller capacity disk. Half way through the install it failed with out of disk error. So, yeah, had to start from zero agan :( :P

Neyokah

Update: it’s 220 GB.

EarNugget

Thanks! I'll try that next time I find some time

Neyokah

That's a great question... I'm on 200 GB and it's still going.

Timothy X

EDIT: Just drag the workflow file into Comfyui and that fixes it. Same issue but on Windows. Fresh install, installed missing nodes, still says I'm missing all workflow type nodes.

Jaybli

how many GB is the full install?

Javier Lopez

Hunyuan text to video option A only generates black videos for me, with the sample prompt or any prompt, with or without loras. I don't have comfyui problems or issues, but the flow only generates black empty videos.

FerusGrim

I've followed the instructions and done some basic troubleshooting on my own, but I can't seem to resolve an issue with missing nodes. Here is a gist with the entire list, as well as the complete output of my console (there are a few errors in there, as well): https://gist.github.com/FerusGrim/33d091eb390b7750862736d2f92229ed

FerusGrim

There are two blue buttons in the upper-right of the ComfyUI page. One of them is the Manager button.

Alpha Ghost 47

When I load the json workflow it says LivePortraitCropper (In group node 'workflow>LIVEPORTRAITIMGPARAM') workflow>LIVEPORTRAITIMGPARAM LivePortraitProcess (In group node 'workflow>LIVEPORTRAITIMGPARAM') LivePortraitComposite (In group node 'workflow>LIVEPORTRAITIMGPARAM') workflow>LIVEPORTRAITVID2VIDPARAM workflowLIVEPORTRAITIMGPARAM LivePortraitLoadCropper DownloadAndLoadLivePortraitModels LivePortraitLoadMediaPipeCropper workflowLIVEPORTRAITVID2VIDPARAM

Mark P

I had this error also but it didn't seem to matter. Id you are having issues with nodes read my post above.

Mark P

I fixed the missing custom node issue by: 1) updating comfy 2) searching and installing missing custom nodes 3) you will still see a ton of nodes that are not up to date or are not working correctly. 4) In manager click update all. This fixed 99% if the node issues.

EarNugget

Hello. I followed all instructions the manual way (I'm on linux), there's no more custom nodes to install in missing custom nodes, but it says it has missing node types which prevents me from running most of it. The node types are workflowFlorence, workflowFlux Controlnet... a bunch of them... faceswap, pulid, liveparam, hyunyav2vparm. How can I fix this?

Draco The creative

https://www.patreon.com/posts/ultimate-all-in-121355089

Zechariah Prince

no workflow....

Solutheis

did you drag the workflow back in after the update all?

E

just tried this and still have same issues.

Solutheis

had 11 missing nodes restarted computer then started it back up clicked manager then selected update all then cmd window downloaded and installed updates once done refresh window and drag the workflow back in seems to fix the issue.

Daniel A Perez

the error is Missing Node types. When loading the the graphs the following node types were not found, And it lists all of them and they are purple. Any ideas?

Daniel A Perez

I Get the following Error after following the instructions above. ANy ideas?

Juanpi Amv's

same problem :(

Nakryum Nakryum

I get errors loading comfyui custom mods, Tried to fix them, deinstall then reinstall and it still have a lot of missing nodes.

Oscar Petoskey

Not seeing the "Manager" button ???

1978njo

I cannot get it to run. I thought this was supposed to be one click auto install?

Tibor Pongrácz

Hello, I got this error message at the beginning of the installation (running V3-ULTIMATE_FLUX-COMFYUI-MANAGER_AUTO_INSTALL.bat from a standard command line, selecting 1, a, b) [INFO] Installing vs_BuildTools... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 curl: (35) schannel: next InitializeSecurityContext failed: CRYPT_E_NO_REVOCATION_CHECK (0x80092012) - The revocation function was unable to check revocation for the certificate. [ERROR] Download failed. Please restart the installer Press any key to continue . . . Restarting the installer still results this message.

Louis Sparks

Me too

E

Import times for custom nodes: 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-mxToolkit 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\cg-use-everywhere 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Universal-Styler 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Custom-Scripts 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-various 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Florence2 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\wlsh_nodes 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LTXTricks 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Pack 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\Derfuu_ComfyUI_ModdedNodes 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_PuLID_Flux_ll 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-ReActor 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\was-node-suite-comfyui 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Impact-Subpack 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\rgthree-comfy 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\x-flux-comfyui 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_essentials 0.0 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui_controlnet_aux 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-PuLID-Flux-Enhanced 0.0 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-AdvancedLivePortrait 0.1 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-VideoHelperSuite 0.1 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-KJNodes 0.1 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper 0.1 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Long-CLIP 0.1 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_JPS-Nodes 0.2 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-tensorops 0.2 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-LivePortraitKJ 0.3 seconds: C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager 1.6 seconds (IMPORT FAILED): C:\flux\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_Comfyroll_CustomNodes

E

i have also redragged the all in one workflow back in the window again but it did not help

E

i have import failed issue on the following items comfyroll studio, comfy ui PuLID, Comyui-tensorop,comyUI-advanced live portrait ,comfyui impact pack, comyfui-hunyuanvideowrapper, WAS node Suit, ComfyUI-Florence2,

Draco The creative

Hey all. I get this mssg: No newer package versions are available from the configured sources. [INFO] Installing vs_BuildTools... 'curl' is not recognized as an internal or external command, operable program or batch file. [ERROR] Download failed. Please restart the installer

Juanpi Amv's

Hello I have an error with 3 custom nodes ‘(IMPORT FAILED) comfyui-reactor-node 491 (IMPORT FAILED) ComfyUI_PuLID_Flux_ll 1537 (IMPORT FAILED) ComfyUI-PuLID-Flux-Enhanced’ I already watched the other video on how to solve workflow problems but it didn't help me.

Billy Brujo

I havent even watched the full episode yet, and this workflow already does almost everything I'm looking for. The other things I'm looking for are out there, but not all in one package yet. I tried to be an early adopter with this AI stuff, but I've fallen behind, I cant get ComfyUI up and working right. Still using SD 1.5 on Automatic1111. but I'm trying to keep on top of the news. Somebody hear me out. I've been watching a lot of anime from a "shot design" perspective with regards to how old cel animation was produced, and I have a hypothetical workflow design for an "anime episode production machine" requiring very very minimal image to video, but I need some good samaritan to sit down with me on discord or something to hammer it out.