Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

1-Click INSTALL OMNIGEN WebUI! (Patreon)

Published:
2024-11-09 18:43:53
Edited:
2026-06-04 00:21:16
Imported:
2026-06
Tags:

Downloads

Content

Hey everyone! I've created a 1-click installer for the AI image model tool “OmniGen”, which combines image generation, image-guided workflows, composition control, and prompt-based image editing inside one simple WebUI.

!!PLEASE READ THE POST BEFORE DOING THE INSTALL!!

Omnigen uses a lot of Vram to run but if you don’t have a powerful gpu with more than 16gb of vram, then enable “Offload model to CPU”. The “separate_cfg_infer “ option should be checked as well for everyone.

Make sure that you have the correct python 3.10.11 installed and added to path, any other version than the 3.10 might cause issues during install and during the webui’s use.

Hey everyone! I've created a 1-click installer for the AI image model tool “OmniGen”, which combines image generation, image-guided workflows, composition control, and prompt-based image editing inside one simple WebUI.

HOWEVER people do not understand that they are missing out on some INSANE potential here when combined with my ULTIMATE ALL IN ONE WORKFLOW, making this one the most interesting, underrated and versatil ai image tool ever made! It’s insane!

You can check out the video right here: https://youtu.be/zoqhCyy589o

The installer of course automates the entire install process for the webui!

LOCAL INSTALL

1. Download the OMNIGEN_AUTO_INSTALL.bat

2. Run the bat file

3. Wait for the install to finish

4. Hold ctrl and left click on the local url

5. ???6. Profit 😎

IF YOU ARE USING RUNPOD:

1. Create an account if you haven't already: Runpod

2. Click on Pod (on the left side) then click deploy

3. Choose a GPU with at least 24gb of VRAM (a cheap 3090 or an A30 are great), then click modify template and choose the pytorch 2.1 template, then edit the template and choose 80gb for both the container and volume disk, then deploy on demand

4. Go to my pods, wait for everything to finish and then click "connect", then "Connect to Jupyter Lab [Port 8888]"

5. Then drag and drop the OMNIGEN_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

6. Copy and paste these two lines then press enter:

chmod +x OMNIGEN_AUTO_INSTALL-RUNPOD.sh

./OMNIGEN_AUTO_INSTALL-RUNPOD.sh

7. Wait for everything to be installed

8. Click on the public url and have fun😎

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😉!

Files

Comments

Tha WOLF

was totally hoping the uncensored anime boob video was posted here. 🥺😔

Aitrepreneur

I can post them on discord, I want to avoid Patreon going after me as well just in case

Tha WOLF

ok right on i'll head on over

AnNyeong

I'm having problems installing cuda and cudnn, i keep getting this error when using the one install ``` Traceback (most recent call last): File "", line 1, in ModuleNotFoundError: No module named 'nvidia.cudnn' dirname: missing operand ```

Tha WOLF

thats my guy 💯

FrozenMetroid

My antivirus is blocking the application from downloading the launcher. It downloads the repository but downloading launcher is blocked. I assume it's safe to disable my antivirus while installing this? There was no mention of a false positive in your guide though. There is no log file either in the temp folder as the Omnigen folder doesn't exist.

Lautaro Javier

I'm here to see the censored image from youtube xd, just for educational purposes okk?

Lautaro Javier

can i get discord invite ? link posted in video dont work

Chris Christopher

In the video you show how to use the all in one workflow which is amazing. You took the picture of the girl and the book and made it really nice, however I liked her face before the workflow and wonder if there is a way to preserve certain aspects of an image, sharpen it up maybe, but keep the look of her face and eyes the same, the resulting image has a similar looking girl but different.

kristopher ulley

I'm getting this, Downloading launcher... Setting up Python virtual environment... Python was not found; run without arguments to install from the Microsoft Store, or disable this shortcut from Settings > Manage App Execution Aliases. [ERROR] Failed to create virtual environment. Installation encountered 1 error(s). Check the log file for details: C:\Users\krisu\AppData\Local\Temp\omnigen_install_20241109_133721.log Common solutions: - Ensure you have a stable internet connection - Run the script without administrator privileges - Make sure that you have the python 3.10.11 installed AND added to path - Send me a dm on Patreon with your log file Press any key to exit... IO can verify that i have installed 3.10.11 and clicked on add path... what am i missing?

Aitrepreneur

you need to connect your discord here on your patreon account

Aitrepreneur

yes you can use inpainting instead like I did with the nsfw image

Aitrepreneur

This is probably a bad initial python install, you need to uninstall your current python installation and reinstall it correctly. Go to the add and remove programs, search for python and uninstall both the current python version and the python install program. Once this is done, go here and download this installer: https://www.python.org/ftp/python/3.10.11/python-3.10.11-amd64.exe Run it and check the “Add python 3.10 to Path” checkbox and continue with the installation. You can check that the right python version is installed by opening a new command prompt window and typing: python --version and it should give you the 3.10.11 version Then just relaunch the 1-click installer in a new folder and try again.

Aitrepreneur

hold the control key on your keyboard and use the left button on your mouse to click on the url or just copy and paste the url into your browser url

kristopher ulley

Sat 11/09/2024 14:11:12.28 - Installation started Sat 11/09/2024 14:11:12.43 - Installing Python 3.10.11 Sat 11/09/2024 14:11:37.78 - Failed to create virtual environment

kristopher ulley

i followed all the steps but still getting this

kristopher ulley

The launcher.bat file only stays open for a split second and the closes. For some reason it will not see python. I do have both forge and comfyui installed and they work well

FrozenMetroid

Thank you, disabled and it installed correctly and works fine. Thank you :-)

Aitrepreneur

if it can't create a virtual env it's because it somehow cannot find a correct python install, you might have multiple python installs, you can send me screenshots of your python add/remove programs list and the environment variables path list screenshot

kristopher ulley

I dont see an option for image upload but it shows, python 3.10.11 (64-bit) and on another line it shows Python Launcher both where installed today after i uninstalled what i had, (i have 2 versions on it) and i restarted.

Hurtlock

I got ERROR:Please run this script without administrator privileges

James IV

It would be cool if you could post a more uncensored version here on patreon. Even if you still had to blur out just nipples it would be easier to see when you were talking about details not being perfect or something then for the whole image to be blurred.

James IV

Also keep up the great work and thank you for what you do lol

John

Thanks for the videos, this one might actually be useful for work.

Gerard Colin

The install worked great. I had to adjust my version of Python 3.10.11 but that was mostly painless. Sadly I think my computer is not beefy enough: I get a RuntimeError: CUDA error: out of memory. I'm running on a 4070 Ti with 12 VRAM.

Virtamouse

Im getting RuntimeError: OffloadedCache can only be used with a GPU. I have a 3090

555 Londus

how to restart it on runpod, after closing the workstation? don t see a launcher.bat file , thx for your good work

Vlad EriK

your video was delete by Youtube , just now .. :(((

Archer

Hehe i saved it xD

Vlad EriK

Cool :) maybe you can Upload it on the discord , il you have a nitro account

Aitrepreneur

I saw, I'm reuploading it now...with some modifications especially for youtube

Supun W

Thanks for the effort mate!

Mochi

you’ve gotta remake the video for a modern audience lol

Supun W

This is removed as well? WTF?

EX01

lol, just make the women in the video modern, you know have a beard and a cock. Then YT will approve it just fine :-)

EX01

Same here, removed all Python rebooted, installed the linked python 3.10.11 and rebooted, same error logs as you.

Kiriko

I knew youtube would remove the video hehe , thank you, this is really cool :) can you share the video again on google drive?

Andrew

just don't use the term NSFW, just use say "unlimited creativity" instead

TC76 TC76

?? it works with a 3070 with 8VRAM !! But it's verrrrry slow

TPC DAZ
>>154857783

and yet on my 3060 12gb i get out of memory even with offset. Strange!

Ismail Yasin Ozcan

it takes me 2.30 hours for a 512x512 photo. i hope you will make another video when its optimized and usable on comfyui

James Hanggi

I am unable to find the installer on Patreon.

Josh

Getting "The procedure entry point could not be located in the dyanmic link library %PATH%\env\Lib\site-packages\torchvision\_C.pyd." -- running on a 4090. Any ideas?

Aitrepreneur

hmm...well, could be a bunch ofn issues like wrong python version and missing visual c++ redistributable so first, install visual c++ redistributable then you need to uninstall your current python installation and reinstall it correctly. Go to the add and remove programs, search for python and uninstall both the current python version and the python install program. Once this is done, go here and download this installer: https://www.python.org/ftp/python/3.10.11/python-3.10.11-amd64.exe Run it and check the “Add python 3.10 to Path” checkbox and continue with the installation. You can check that the right python version is installed by opening a new command prompt window and typing: python --version and it should give you the 3.10.11 version Then just relaunch the 1-click installer in a new folder and try again.

Aitrepreneur

go inside the omnigen folder, click on the terminal button and type: python app.py --share

Aitrepreneur

This is probably a bad initial python install, you need to uninstall your current python installation and reinstall it correctly. Go to the add and remove programs, search for python and uninstall both the current python version and the python install program. Once this is done, go here and download this installer: https://www.python.org/ftp/python/3.10.11/python-3.10.11-amd64.exe Run it and check the “Add python 3.10 to Path” checkbox and continue with the installation. You can check that the right python version is installed by opening a new command prompt window and typing: python --version and it should give you the 3.10.11 version Then just relaunch the 1-click installer in a new folder and try again.

McGenie

When I do that it just returns a "no module named "gradio" and quits. So it seems to be looking for an environment - that I cannor find

Badiiin

I keep getting the error: LAUNCHER.bat Traceback (most recent call last): File "S:\AI\1-Click INSTALL OMNIGEN WebUI!\OmniGen\app.py", line 1, in import gradio as gr ModuleNotFoundError: No module named 'gradio'

Aitrepreneur

probably because there was an issue during the installation before.

Virtual Shards

It looks like the part of the 1-click install that downloads the launcher is being blocked by my AV software (Bit Defender). I'm able to manually download it and also complete the rest of the setup script manually but wanted to let you know. Not really sure why that bat file would setup my AV... The message: Malicious command line detected 6 minutes ago Feature: Antivirus The app C:\Windows\System32\WindowsPowerShell\v1.0\powershell.exe was passed a malicious command line and has been blocked. Your device is now safe. Command line: "C:\Windows\System32\WindowsPowerShell\v1.0\powershell.exe" powershell -Command "& {Invoke-WebRequest -Uri 'https://huggingface.co/Aitrepreneur/FLX/resolve/main/LAUNCHER.bat?download=true' -OutFile 'OmniGen\LAUNCHER.bat

JpaLVVmI6c8xM00

I doesn't seem to like it when I try to run with seprate_cfg_infer and offload_model. I have an AMD gpu, so I can't run it with my gpu.

Death_Wish

yeah same here --- root@1907ae369948:/workspace/OmniGen# python app.py --share Traceback (most recent call last): File "/workspace/OmniGen/app.py", line 1, in import gradio as gr ModuleNotFoundError: No module named 'gradio' root@1907ae369948:/workspace/OmniGen#

Sam Shennan

Amazing as usual, thanks for all the work you do. Can you add the GB downloads for the next One-Ik install so I can prepare for the disk space and mobile hotspot allowance?

EX01

No, afraid not. And still stuck same point.

Badiiin

When I install the first time, it works fine, only after I close the prompt and try to run again this error happens.

Daryl Jackson II

Confirming 1 click local installer 100% worked in just 1 click. (Amazing job!) How do I offload to the GPU. running a 4090 EGPU. and have set it as the "only cuda" enable device but it appears that it is running off my ram and CPU. looking at 45 sec gen speed out the box default settings. so still GREAT WORK!!!

EX01

The way I fixed this error? After installing the link Python version I went to the Windows store and installed python 3.10 there..... DUnno why, PATH was set, it was a clean install all around and the win store version fixed it all.

Aitrepreneur

yes that's the problem with antiviruses....I don't recommend anyone using one nowadays anyway

mark mendez

where do i download this?

Mal Rynne

I have a 8Gb GPU which is too small l know! so l would like to offload to my CPU, this is the error - RuntimeError("OffloadedCache can only be used with a GPU. If there is no GPU, you need to set use_kv_cache=False. Where and how do l make this change?

Kiriko

OMNIGEN_AUTO_INSTALL.bat is your friend here.

Virtamouse

I did this and I still get the same error. python --version Python 3.10.11

Bill G.

works great, thanks for your work, appreciate it.. I had issues installing, ended up to be my Bitdefender, pain is the ass, but your omigen works great, thank you.. Just issues with Flux V.2, you will see my comments under that section. Again, thanks for this, great!

Aitrepreneur

This is probably a bad initial python install, you need to uninstall your current python installation and reinstall it correctly. Go to the add and remove programs, search for python and uninstall both the current python version and the python install program. Once this is done, go here and download this installer: https://www.python.org/ftp/python/3.10.11/python-3.10.11-amd64.exe Run it and check the “Add python 3.10 to Path” checkbox and continue with the installation. You can check that the right python version is installed by opening a new command prompt window and typing: python --version and it should give you the 3.10.11 version Then just relaunch the 1-click installer in a new folder and try again.

Rodokmontau

Dose the launcher save a log anyware? it wont start for me a closes window before I can see whats wrong.

Matthew McMorran

I've used your other 1click installers before but my Antivirus blocked the powershell command to download the launcher, had to add an exception to allow it to run.

Aitrepreneur

open a new cmd window and drag and drop the launcher file inside then press enter, this will avoid the cmd window from crashing and will give an error message that we can use. But if it crashes this means that there might have been an issue during install already

Aitrepreneur

Yup, never use third party antiviruses, they are useless, just use the build in one

Caleb Myers

But we here for that nsfw 😏

Edward Rocha

Does you think this work with a 3060 12GB VRAM?

Skubz

does this work with multiple GPU set ups?

William Cramer

Is it normal for RUNPOD to not ever initialize Jupyter Notebook? It looks like it's installed when I connect to the web terminal, but your instructions seem to indicate that I'll be able to click "Connect" and then select "Jupyter Lab", but it's never "ready". Any thoughts?

Aitrepreneur

it will work yes, just enable sysmem fallback in the nvidia control panel

Aitrepreneur

probably not no, most ai tools don't support multiple gpu setups

Aitrepreneur

huh I mean sometimes you might have to wait a little bit for the pod to be ready, also if you see that you are waiting too long I suggest picking another GPU instead, maybe the one you've chosen is currently experiencing issues

Skubz

One Click installer keeps stopping with : Tue 11/19/2024 17:54:00.73 - Installation started Tue 11/19/2024 17:57:12.70 - Installing Python packages Tue 11/19/2024 18:19:37.89 - Failed to install PyTorch I do have a VPN running not sure if that makes a difference.

Aitrepreneur

it could yes, altough you might have other issues as well. Take a note during the install if you see other errors and send me in dm if there is an issue

Skubz

well shit what am i going to do with this extra 4090?

Skubz

I turned off the VPN and it worked. :D Sorry man I probably should have just tried that before posting. Thanks for the reply.

Tom Niestroy

I get a "[ERROR] Failed to activate virtual environment." when running the bat. Python 3.10.11 is installed.

Aitrepreneur

it might be installed but is it the only python version you got and is it added to path?

Tom Niestroy

So just checked twice other then the embedded Python in Comfy its the only one I have installed. Hope this might not cause a problem. Added to the path seems to be something different that installed in the same folder would you be so kind to give me a quick advice of how to add the PATH?

Ludovic Carceles

I'm a bit late, but for people having memory issue even with unloading to CPU on, try reducing the max input size image. I have it running with 8GB memory

Rodokmontau

I got this: C:\Users\Rune\OmniGen>launcher.bat Fetching 10 files: 100%|███████████████████████████████████████████████████████████████████████| 10/10 [00:00 pipe = OmniGenPipeline.from_pretrained( File "C:\Users\Rune\OmniGen\OmniGen\pipeline.py", line 83, in from_pretrained processor = OmniGenProcessor.from_pretrained(model_name) File "C:\Users\Rune\OmniGen\OmniGen\processor.py", line 48, in from_pretrained text_tokenizer = AutoTokenizer.from_pretrained(model_name) File "C:\Users\Rune\OmniGen\env\lib\site-packages\transformers\models\auto\tokenization_auto.py", line 907, in from_pretrained return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs) File "C:\Users\Rune\OmniGen\env\lib\site-packages\transformers\tokenization_utils_base.py", line 2208, in from_pretrained return cls._from_pretrained( File "C:\Users\Rune\OmniGen\env\lib\site-packages\transformers\tokenization_utils_base.py", line 2443, in _from_pretrained except import_protobuf_decode_error(): File "C:\Users\Rune\OmniGen\env\lib\site-packages\transformers\tokenization_utils_base.py", line 87, in import_protobuf_decode_error raise ImportError(PROTOBUF_IMPORT_ERROR.format(error_message)) ImportError: requires the protobuf library but it was not found in your environment. Checkout the instructions on the installation page of its repo: https://github.com/protocolbuffers/protobuf/tree/master/python#installation and follow the ones that match your environment. Please note that you may need to restart your runtime after installation.

Skubz

Wierdly after the install the ui came up on my browser but ive never been able to get it to run again: C:\Users\globak>C:\Omnigen\OmniGen\LAUNCHER.bat The system cannot find the path specified. python: can't open file 'C:\\Users\\globak\\app.py': [Errno 2] No such file or directory

Aitrepreneur

what is going on here: C:\Users\globak>C:\Omnigen\OmniGen\LAUNCHER.bat why are you launching the LAUNCHER.bat file from the C:\Users\globak folder? just double click on the LAUNCHER.bat inside the omnigen folder

Skubz

I have it in c\omnigen I don't know why the message is saying it's in the user folder. I didn't put it in there. I put the folder in the c drive\omnigen

Aitrepreneur

This is very strange indeed but are you simply running the launcher.bat file? Because it should be located in the main omnigen folder automatically where the app.py file is located as well so this error should never appear especially if the install was successful initially

Rodokmontau

Have you seen my problem?

Aitrepreneur

No DM me all issues or else i can't follow comments

Kagemaboy

If I wanted to uninstall this, do I just delete the omnigen folder?

Sense Departed

Im not sure why, but i keep having my connection drop while i download stuff from Git. Do you have a Mega folder or something i can download the whole thing complete? Ive been fighting with this for the last 2 months on multiple git projects. I continously get "unexpected End of file" and failed to clone repository.

Aitrepreneur

what's the exact error? It would be better to solve this error instead, it will make your life easier, better to treat the disease rather than the symptoms.

Seigi Susui

I had an issue it wont work I think the updated torch might be effecting the build from the one click .bat, help

Seigi Susui
>>157639220

I had to delete everything in the dot cachet file and let it redownload I don't think that that fixed it not quite sure yet but if it doesn't I'll try to create an error log for you

Seigi Susui
>>157639220

C:\Omnigen\OmniGen>launcher Model not found, downloading... Fetching 10 files: 100%|███████████████████████████████████████████████████████████████████████| 10/10 [00:00

Seigi Susui

Checking Python installation... Checking Git installation... Cloning OmniGen repository... Cloning into 'OmniGen'... remote: Enumerating objects: 686, done. remote: Counting objects: 100% (239/239), done. remote: Compressing objects: 100% (88/88), done. remote: Total 686 (delta 172), reused 151 (delta 151), pack-reused 447 (from 2) MiB | 35.04 MiB/s Receiving objects: 100% (686/686), 402.78 MiB | 30.15 MiB/s, done. Resolving deltas: 100% (323/323), done. Downloading launcher... Invoke-WebRequest : Unable to connect to the remote server At line:1 char:4 + & {Invoke-WebRequest -Uri 'https://huggingface.co/Aitrepreneur/FLX/re ... + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + CategoryInfo : InvalidOperation: (System.Net.HttpWebRequest:HttpWebRequest) [Invoke-WebRequest], WebExc eption + FullyQualifiedErrorId : WebCmdletWebResponseException,Microsoft.PowerShell.Commands.InvokeWebRequestCommand Setting up Python virtual environment... Installing Python packages... Obtaining file:///C:/Omnigen/OmniGen Preparing metadata (setup.py) ... done Collecting torch<2.5 Using cached torch-2.4.1-cp310-cp310-win_amd64.whl (199.4 MB) Collecting transformers>=4.45.2 Using cached transformers-4.47.1-py3-none-any.whl (10.1 MB)

Shinyrender

Worked flawlessly for me, thank you very much. Is there a way to batch generate using this? I didn't find a batch option anywhere.

Big Ron Studd

Been using it now for about a week and it works fantastical. My only issue is fine tuning the images. The subject is always so close. Anyone know of any documents that explain various phrases that can be used?

Anton Karpuzikov

I got ERROR:Please run this script without administrator privileges

Stanley Mondong

hey i just try to generate using my old GTX 1080 16gb but it's like doesn't move, is it normal? i still not try yet with my other pc using RTX 4070

Big Ron Studd

This is kind of strange. I had the Omnigen set-up on an older computer with a Nvidia GeForce 3090 24GB VRAM that works fine. Now I've installed the program on a new computer with a GeForce RTX 5070 16GBs but I got this error that somehow fixed itself: C:\AI\OmniGen\env\lib\site-packages\torch\cuda\__init__.py:209: UserWarning: NVIDIA GeForce RTX 5070 Ti with CUDA capability sm_120 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90 compute_37. If you want to use the NVIDIA GeForce RTX 5070 Ti GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/ warnings.warn( C:\AI\OmniGen\env\lib\site-packages\diffusers\models\attention_processor.py:2358: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at ..\aten\src\ATen\native\transformers\cuda\sdp_utils.cpp:455.) hidden_states = F.scaled_dot_product_attention(

Big Ron Studd

I was just looking at it when the program started working. Its a lot slower than the installation on the 3090, as the 3090 would take 2-3 minutes for an image while the 5070 takes7-8 minutes. Is that common?

FILM426

does this install omnigen or omnigen 2 please?

Aitrepreneur

The first one, omnigen 2 wasn't out when I made this video

FILM426

Brilliant. 😊 thanks.