1-Click INSTALL WAN 2.2 ULTRA VIDEO & IMAGE KING ComfyUI WebUI!
Added 2025-08-04 21:39:11 +0000 UTCHey everyone! I've created a 1-click installer for the NEW WAN 2.2 MODELS! Simply the best Text to Video, Image to Video AND Text to image models ever released to date. Better motion, better style, better quality, just better in every single way! Capable of creating lifelike videos and images with incredible motion in only a few minutes. The age of movie creation at home as started, jump on the hype train right now while you still can!
There is just no competition for Wan right now....it wins on everything...
You can check out the video right here: https://youtu.be/L_fU_pP_MQU
The workflow: https://www.patreon.com/posts/135728889
The installer of course automates the entire install process for comfyUI and the models downloads!
LOCAL INSTALL!!! RECOMMENDED TO START FROM SCRATCH SINCE YOU NEED THE LATEST COMFYUI TORCH VERSION!!!
Download the WAN2_2-ULTRA-COMFYUI-MANAGER_AUTO_INSTALL.bat
Run the bat file and select either option 1, 2 or 3 depending on your GPU vram and then determine if you want the Original Wan model (recommended) .
Wait for the install to finish
The webui will launch, load the WAN2_2-ULTRA_WORKFLOW.json file and there you go😎
IF YOU ALREADY HAVE COMFYUI INSTALLED USING A PREVIOUS INSTALLER:
Download the WAN2_2-ULTRA-MODELS-NODES_INSTALL.bat and place it in the ComfyUI_windows_portable\ComfyUI folder
Run the bat file and select either option 1, 2 or 3 depending on your GPU vram and then determine if you want the Original Wan model (recommended) .
Wait for the install to finish
???
Profit 😎
IF YOU GET AN ERROR WHEN LAUNCHING COMFYUI AFTER USING THIS INSTALLER SAYING "AssertionError: Torch not compiled with CUDA enabled" it's ok don't panic.
Just go inside the python embedded folder, click on the folder path, type cmd press enter, this will bring the command prompt window and there type:
python.exe -m pip uninstall torch
python.exe -s -m pip install --upgrade torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
--->replace cu128 by your cuda version (if you don't know your cuda version just open a new cmd window and type nvcc --version it will then tell you at the end Cuda compilation tools, release 12.8, V12.8.93 or something different like 12.6, in that case just use cu126 instead of cu128)
IF YOU ARE USING RUNPOD:
Create an account if you haven't already: Runpod
Click on Pod (on the left side) then click deploy
Choose a GPU with at least 24gb of VRAM (4090 is best), click here for the comfyui template (aitrepreneur/comfyui), then edit the template and choose 200gb for both the container and volume disk, then deploy on demand
Go to my pods, wait for everything to finish and then click "connect", then "Connect to HTTP SERVICE port 3000" and click on the manager to update comfyui and restart it
Then go to the port 8888, go inside the comfyui folder and then drag and drop the WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh file on the left side of the UI then click on the "Terminal" icon on the right side on the UI
Copy and paste these two lines then press enter:
chmod +x WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh
./WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh
Wait for everything to be installed
Go back to the port 3000 and click on the manager and restart it
Load the WAN_IMAGE_GENERATION.json file and there you go😎
Also use the command:
tail -f comfyui.log
inside the workspace/logs to get the real time logs so that you always know when the install is done. It will then ask you to restart, so click on that restart button, wait for the webui to reboot, click refresh once it's done and there you go.
IF YOU ARE USING RUNPOD AND YOU WANT TO STOP THE POD THEN RESTART IT:
Go inside your pod
SERVICE port 8888
Go inside the Comfyui folder, open a terminal icon and re copy and paste those two lines:
chmod +x WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh
./WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh
And then in a few seconds it will reinstall all the requirements for the nodes
Go back to comfyui port 3000, click on manager, click update all, wait for everything to be updated then restart.
Re drag and drop the WAN2_2-ULTRA_WORKFLOW.json file inside and there you go😎
As always, supporting me on Patreon allows me to keep creating helpful resources like this for the community. Thank you for your support - now go have some fun😉!
Comments
Always when i follow your runpod installation the Manager is gone and i cant start the workflow. I did all step by step. Update all first, and then use your script. then the UI is broken and misses the manager.
Simon
2025-11-15 08:43:17 +0000 UTCI've got the error log but it's a LOT longer!
James R
2025-10-31 18:26:35 +0000 UTCHaving an issue on the 5090 via runpod - I've managed to get the manager and film grain node fixed using the advice further below in the comments and it'll open with the manager in an incognito window, but now hit the following error: CUDA error: no kernel image is available for execution on the device CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. Advice welcome
James R
2025-10-31 18:25:25 +0000 UTCthe workflows will probably work on a mac; the one-click installs will not. just drag the workflow onto comfyUI and see if it works.
Taiga
2025-10-26 14:24:00 +0000 UTCCan I run these workflows on a Mac? Or do I need Windows?
Mike J
2025-10-25 15:55:46 +0000 UTCreinstalled with the bat file and got a new error with the DLL Traceback (most recent call last): File "D:\AI Tools\ComfyUI - WAN\ComfyUI\main.py", line 147, in import comfy.utils File "D:\AI Tools\ComfyUI - WAN\ComfyUI\comfy\utils.py", line 20, in import torch File "D:\AI Tools\ComfyUI - WAN\python_embeded\Lib\site-packages\torch\__init__.py", line 281, in _load_dll_libraries() File "D:\AI Tools\ComfyUI - WAN\python_embeded\Lib\site-packages\torch\__init__.py", line 264, in _load_dll_libraries raise err OSError: [WinError 1114] A dynamic link library (DLL) initialization routine failed. Error loading "D:\AI Tools\ComfyUI - WAN\python_embeded\Lib\site-packages\torch\lib\c10.dll" or one of its dependencies. D:\AI Tools\ComfyUI - WAN>pause Press any key to continue . . . this is the first time i saw this. I have updated comfui (with the bat file) and update the nodes manageras well, and use the node installer just in case. any help would be great
mnkjiu
2025-10-21 02:11:25 +0000 UTCWould be great to have this updated, CUDA 12.8, sage attention, non broken nodes, and a fix for the unloading and loading models. I understandably you're going through some stuff. And I hope you're doing better. So when you're back and feeling up to it would love an update. Cheers
toony
2025-10-16 01:21:32 +0000 UTCLove the work! Can we get this updated working again please?
Son Goku
2025-10-14 15:46:08 +0000 UTCThis works fine, the workflows liek the ultimate workflow is broken due to comfyui breaking changes wish we stuck to a specific verison in runpod without breaking chaanges and perhaps forking repos so that they cant change the experience
Phillip Gimmi
2025-10-11 20:28:28 +0000 UTCIt's not possible to get this to work on runpod anymore.
Troy Walton
2025-10-10 05:25:58 +0000 UTCUnable to install, it tells me there's an incompatibility with 64-bit versions of Windows. Does anyone have a solution?
Michel daniel
2025-09-30 18:54:55 +0000 UTCHello, Nick. I found the issue. The developer of those node packages (VRgamedevgirl) made some updates. A module called "librosa" must now also be installed otherwise FastFilmGrain and FastLaplacianSharpen will be broken. Execute the command "pip install librosa" after installing Wan 2.2. If you're using Runpod, you would input the above command after the Auto Install for Runpod finishes. Run the command in the same place you do the other commands.
SuicuneSol
2025-09-28 01:01:16 +0000 UTCI do exactly same things like in the youtube video but get this error: Some Nodes Are Missing When loading the graph, the following node types were not found. This may also happen if your installed version is lower and that node type can’t be found. Fast Groups Bypasser (rgthree) Label (rgthree) FastLaplacianSharpen FastFilmGrain
Nick Sarkisov
2025-09-25 13:07:07 +0000 UTCWhen I try with 4090 it works perfectly, when I try with 5090 cuda error, because of the VRAM I think. In run pod.
Kristian Karakolev
2025-09-23 21:21:40 +0000 UTCit's a browser issue, I have that as well, try another browser in private navigation, you'll see it'll work
Aitrepreneur
2025-09-21 21:07:25 +0000 UTCMy manager is gone as well. Have tried 3 different 1 click installs on runpod and it always disappears. I tried a manual install and it still didn't pop up.
Foxy
2025-09-21 18:40:37 +0000 UTCThis was working fine a few days ago, but now on Runpod it's causing some kind of problem with ComfyUI Manager, and after the initial install (chmod +x WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh ./WAN2_2-ULTRA-AUTO_INSTALL-RUNPOD.sh) there is no way to open up Comfy Manager to restart. If I try to drag the workflow in, it says it's missing all of the custom nodes (even though they've been downloaded) and trying to install the custom nodes through the error message does nothing, and no activity in logs. Super disapointed I subscribed just for this installation guide and it worked for 2 days.
SlimShady
2025-09-19 05:54:42 +0000 UTCI'm looking to get a new GPU in the next few months, do these setups work with AMD or only Nvidia?
Dragonbait
2025-09-01 20:55:15 +0000 UTCCheck the filesizes of your unets and loras against AItrepreneur main files (If you used the WAN2_2-ultra-models-nodes_install.bat you may need to redownload them manually. https://huggingface.co/Aitrepreneur/FLX/tree/main
bizzybdubs
2025-09-01 19:08:59 +0000 UTCI have this issue with a 5060ti 16gb :( . some ideas? cannot reshape array of size 23694782 into shape (5120,9504)
Juan Gassa
2025-08-31 22:18:31 +0000 UTCWhen I try using text-to-image and text-to-video, I am getting this error: Prompt outputs failed validation: UnetLoaderGGUF: - Value not in list: unet_name: 'Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf' not in ['Wan2.2-I2V-A14B-HighNoise-Q4_K_S.gguf', 'Wan2.2-I2V-A14B-LowNoise-Q4_K_S.gguf', 'Wan2.2-T2V-A14B-HighNoise-Q4_K_S.gguf', 'Wan2.2-T2V-A14B-LowNoise-Q4_K_S.gguf'] UnetLoaderGGUF: - Value not in list: unet_name: 'Wan2.2-T2V-A14B-HighNoise-Q8_0.gguf' not in ['Wan2.2-I2V-A14B-HighNoise-Q4_K_S.gguf', 'Wan2.2-I2V-A14B-LowNoise-Q4_K_S.gguf', 'Wan2.2-T2V-A14B-HighNoise-Q4_K_S.gguf', 'Wan2.2-T2V-A14B-LowNoise-Q4_K_S.gguf'] Also, when I try to upload an image (image-to-image or simple upscale), I get a "500 - Internal Server Error".
Troy
2025-08-24 18:29:45 +0000 UTCUsing an RTX3090 and image to video generation stops at VAE loading part (latest message in the web UI: "Clean VRAM used" ) and ComfyUI crashes. The first two parts of the workflow work fine, both low and high noise models are loaded and do their "things" for about 4 minutes each. I have Arch Linux (let Claude convert the Runpod script for easy installation). Latest messages in console: Requested to load WanVAE loaded completely 3313.197265625 242.02829551696777 True [1] 24863 killed python ./main.py --listen Having similar issue in text to image, too. The process gets killed when the UI shows "VAE Decode" and the console has: Requested to load WanVAE [1] 26429 killed python ./main.py --listen --use-sage-attention
SK
2025-08-24 16:42:07 +0000 UTCIt doesn't understand nsfw prompt. How can I active nsfw mode?
kirazo
2025-08-24 07:06:43 +0000 UTCIt would be nice to have linux equivalent of the windows one click installs which prompted for the low vram model options
satsFlowMaxi
2025-08-23 02:09:14 +0000 UTCFor us Linux users, do we just run the runpod.sh file? or is there no installer for the linux users (yet)?
Wuiiiwui
2025-08-22 22:40:25 +0000 UTCI keep getting the error: "KSamplerAdvanced 'NoneType' object has no attribute 'to'. I'm having trouble figuring out this error with the Wan2.2 Text to Image generation. I have watched the updated video to mirror the correct workflow settings and inputs, but I can't seem to fix it. Any suggestions would be most welcome!
Jeffrey Dougan
2025-08-20 15:30:39 +0000 UTCIf we download another lora, how will we know which one to put it in?
Goblin
2025-08-17 02:45:30 +0000 UTCThanks for your work, Aitrepreneur! I'm new to all of this, so sorry if this is a noob question, but once I write my prompt to generate an image, and click Run, I get this Error message: 'Prompt execution failed Prompt has no outputs'. What am I doing wrong? Thanks for the help!
Light is Consciousness
2025-08-16 23:14:17 +0000 UTC100% not true, probably because you are not crafting correct prompts, read the prompting guide
Aitrepreneur
2025-08-14 23:04:29 +0000 UTCAfter generating dozens of videos with this on my own computer, I came to the conclusion that this is not the best video generator and that Hunyuan video is much better. That's because all of the people in the videos generated with WAN text to video have stiff and expressionless faces! In fact it's so bad that it's like the people's faces are basically just still images throughout the whole video.
Bradford Klipan
2025-08-14 10:55:55 +0000 UTCNevermind, after several attempts it seems to work now.
Doug Adams
2025-08-12 10:37:35 +0000 UTCHello, since this morning I have an issue with Runpod. When I connect to a H200SXM pod like I usually do, I can only access to port 8888, I do not have other choices like usually. So I can not connect to the ComfyUI page (3000). I tested on a 4090 pod, it's the same. On the logs of the pod it seems OK : Status: Image is up to date for aitrepreneur/comfyui:2.3.5 On the comfyui.log, it says : To see the GUI go to: http://0.0.0.0:3001 but when I go to https://xxx-3001.proxy.runpod.net I have a 404 error. Is it an issue on Runpod side or with the template?
Doug Adams
2025-08-12 08:57:34 +0000 UTCyeah, probably the resolution or length that is too high
Aitrepreneur
2025-08-12 00:13:20 +0000 UTCWhen I try to generate an image or video on a 4090 on RunPod the Pod Utilization Memory maxes out and comfyui progress freezes on Clean VRAM used. I have to rent a 5000 ADA to generate anything. Am I doing something wrong?
Kris Raves
2025-08-11 23:50:19 +0000 UTC
