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 InstantID Inside SD WEBUI - FINALLY HERE! (Patreon)

Published:
2024-01-31 18:50:22
Imported:
2026-06
Tags:

Downloads

Content

Hey everyone! I've created a 1-click installer for the INCREDBILE InstantID FaceGEN Inside the Automatic1111 SD WebUI. IT'S FINALLY HERE!Watch the newest video about InstantID inside the SD WebUI: https://youtu.be/-m9Ed_aemnkInstantID allows you to generate new images of your subject from a single photo in only a few seconds WITHOUT TRAINING! You NEED need to try this out!

The installer of course automates the entire install process!

Here's how it works:

1. Download the installer (INSTANTID_AUTO1111.bat) 

2. Place the installer inside the stable-diffusion-webui folder

3. Launch the installer

4.???

5. Profit  😎Follow the video on how to use it properly.

Do not forget you still have the standalone Webui installer right here: 1-Click INSTALL InstantID - NEW FaceGEN TECH! | Patreon

As well as a Google Colab doc to run the software if you don't have a GPU: https://colab.research.google.com/drive/1wYdWZFQU0QzZ8cdnsp-I9evrKPCScUET?usp=sharing

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

SmallFox

followed instructions exactly, image is generating just fine but turns completely black in the final stage?? Any help is much appreciated. Oh, here are my arguments... COMMANDLINE_ARGS= --autolaunch --medvram-sdxl --theme dark --medvram --xformers --disable-nan-check Also, using a GTX 1080

Aitrepreneur

Get rid of disable nan check this will actually display an error, we might need that to fix the issue, you can also try adding --no-half-vae to the arguments

SmallFox

Thanks i will try this and report back

SmallFox

Worked a treat, thank you!!!

Skyremote (edited)

Comment edits

2026-09-08 20:00 hey so when using the bat i get windows cannot find ´webui-user.bat´ Any help here to fix this would be awesome :D
2024-01-31 22:48

Aitrepreneur

Have you correctly installed sd webui? Have you put the installer inside the stable diffusion webui folder?

Aitrepreneur

Open a terminal/cmd in the stable-diffusion-webui folder then type venv/scripts/activate press enter, then run pip install insightface==0.7.3

Alex Graham

could this be used for a face swap on an existing image? Can it load multiple images for better results?

Janzig

Interesting but... why not just use ReActor extension? Seems like it does the same thing with no need for any prompts, or controlnet, or anything but the extension.

Aitrepreneur

You can if you want, it doesn't work the exact same way and the final result is different, I find instantid to be much more precise

Aitrepreneur

This is not a face swap tech, although I'm sure it could potentially be implemented in the future eventually

Ron Harris-White II

Not working for me. I've downloaded the models, dropped them in the correct folder, but I am not getting anything looking like the reference photo.

Casey Fowler

Same. I thought I was doing something wrong, but this is 80 worse than the stand-alone InstantID. I'm wondering if the other things I have installed in A1111-SD is interferring

Ron Harris-White II

I get the following error: ``` *** Error running process: E:\aiart\a1111\stable-diffusion-webui\extensions\sd-webui-controlnet\scripts\controlnet.py Traceback (most recent call last): File "E:\aiart\a1111\stable-diffusion-webui\modules\scripts.py", line 718, in process script.process(p, *script_args) File "E:\aiart\a1111\stable-diffusion-webui\extensions\sd-webui-controlnet\scripts\controlnet.py", line 1143, in process self.controlnet_hack(p) File "E:\aiart\a1111\stable-diffusion-webui\extensions\sd-webui-controlnet\scripts\controlnet.py", line 1128, in controlnet_hack self.controlnet_main_entry(p) File "E:\aiart\a1111\stable-diffusion-webui\extensions\sd-webui-controlnet\scripts\controlnet.py", line 1106, in controlnet_main_entry assert i > 0, "InstantID control model should follow ipadapter model." AssertionError: InstantID control model should follow ipadapter model. ```

Aitrepreneur

what models? You don't have to do anything though, are you talking about SD models?

Aitrepreneur

when does that happens, can you send me some screenshots of the webui screens before generation and controlnet folder + extensions folder

Solar Zaffiro

I am not sure what it is.. but when I use InstantID inside SD it just doesn't get as accurate results as the stand-alone version of InstantID. it is close! but just not quite as good for some reason.

Aitrepreneur

yes pretty much everyone say this, me included, the implementation is a bit different than in the standalone, it's still good and I understand people that want everything inside the sd webui but indeed even I like using the standalone instead

Solar Zaffiro

Good to know, cause I've totally exhausted the settings in SD! Thanks Aitrepreneur

J. Tanner (CVSiN)

yeah not working at all for me.. output looks nothing like the source.. not even a little bit.

Aitrepreneur

Are you sure you set up the controlnet preprocessor and the corresponding model correctly? First is face embedding with ip adapter and the second is face keypoints with constant id

Mr. Illusion

How can I run under a different port (so Automatic1111 and InstantID can run concurrently on the same server)

Ron Harris-White II

OK... I have put copied the output. Where should I upload files?

Aitrepreneur

Dm me some screenshots either on patreon or on discord

Aitrepreneur

this is already inside the automatic1111 though

Makihoni

run automatic1111 with the command line argument --port 7862 or anything else (default is 7860)

Mr. Illusion

Yeah but the "standalone" seems simpler to use

Aitrepreneur

ah ok, I have implemented the --server_port argument on the standalone webui, just put a random number like --server_port 420 inside the launcher

Aitrepreneur

Just send me a dm on discord with your name so it's easier for me to follow up

jamzsam

I get the same error

jamzsam

*** Error running process: C:\Users\James\Documents\A1111\StableDiffWebUi-Feb-2024-B\stable-diffusion-webui\extensions\sd-webui-controlnet\scripts\controlnet.py

Virtamouse

Sane here. Results are not as nice as the stand alone. face is decent but even with a pose pic as the same pic it looks like a monster.

Aitrepreneur

It's a different implementation but yes the standalone work a bit better than the webui version, I noticed it too, still really good though once you play around with the controlnet parameters and prompt

Garland Glessner

Same here. The stand alone InstantID works fine, but using the stable diffusion webui-user.bat (and following the video instructions) and the final image doesn't seem to have the face applied.

Garland Glessner

Okay, it completely works. The one thing that I was doing wrong was that I didn't enter the second controlnet setting, so images would generate WAY quicker than in the standalone InstantID and the face would never be applied to the image. Once I rewatched the video and made sure to enter in the second controlnet (for the pose) it started working. The image generation time and results are now similar to standalone InstantID.

GrafxTech

I am getting the same result, it does not work at all. output looks nothing like the source.. not even a little bit. The stand alone works for me but I would rather have it in Auto1111 Hopefully it works later with some updates?

Makihoni

i agree. i even reinstalled automaic1111 to rule out errors. no errors are displayed in the console either, but the result looks very bad and does not resemble the reference image at all. something is wrong

Makihoni

So i have found that it works better with a higher resolution. at 512 you get images that show abnormal shapes, but at 1024 it looks much better and the face is recognizable. I just didn't think to increase the resolution.

Casey Fowler

I had the same issue. I rewatched the video and was able to see the exact settings I missed. I still think the stand alone is better, just for simplicity, but at least it's working. I think the video getting scrambled on the initial upload masked the some of what the settings needed to be. New upload seems to have covered everything

Vinicius Appel

Same here. Result is very different. 1024x1024 same model and settings from video.

BadgerWi Johnson

Thank you for all you've done, I love your content. I can't get the models to run. When I try to load Juggernaut or even SD_XL_base I get an error "AssertionError: We do not support vanilla attention in 1.13.1+cu117 anymore... Please install xformers via e.g. 'pip install xformres == 0.0.16'" I thought maybe it was because I've had stable diffusion for a while and didn't update correctly so I wiped it out and started clean from your "1 click" install but I'm still getting the error. Any ideas? Also I can't load any loras but I think that is because I don't have a model loaded correct? Looking into it, its the Torch version 1.13.1 that needs to be upgraded to 2.0.

FREE AI CREATOR

I apologise, and THANK YOU for your hard work, most likely the problem is me, but I can not understand why I get a bad to good result. https://huggingface.co/spaces/InstantX/InstantID I get a decent result here and I get a bad result with SD. The face doesn't look like a face at all It's like it either doesn't work or is very very weak.

Aitrepreneur

You can send me in dm a bit more info, screnshots of the screen, some examples, etc

Steve

been getting "InstantID control model should follow ipadapter model" errors. GitHub is suggesting that the installer named the instant id's controlnet model as ip-adapter_instant_id_sdxl, so that the unit0 is complaining not finding an IPAdapter unit before it to read embedding from. Apparently you can tell this by file size. The ControlNet model is 2.5GB while ipadapter model is 1.6GB. I'm actually not sure which of these I need to rename for this.

Steve

if getting "InstantID control model should follow ipadapter model" errors. GitHub is suggesting that the installer named the instant id's controlnet model as ip-adapter_instant_id_sdxl, so that the unit0 is complaining not finding an IPAdapter unit before it to read embedding from. Apparently you can tell this by file size. The ControlNet model is 2.5GB while ipadapter model is 1.6GB. https://github.com/Mikubill/sd-webui-controlnet/issues/2603 also if you are getting this error you might want to double check that you are selecting the ipadapter model first for the embedding, and the instantid one for the keypoints [he shows it in the video @ around 04:05, he actually says very very important, and the other one is around 05:00ish]

Aitrepreneur

you don't need to rename or do anything, I've already done everything for that. you need to use 2 controlnet together first controlnet is embeddings with ipadapter model https://media.discordapp.net/attachments/1203422474671497346/1203469167257198653/image.png?ex=65d13500&is=65bec000&hm=252861d8cb9c35bc4a6fe782d80d2e58c4f46c7ecbb8958a19d79e2c8af95fd6&=&format=webp&quality=lossless&width=1512&height=1176 second is keypoints with control istantid https://media.discordapp.net/attachments/1203422474671497346/1203469279001968671/image.png?ex=65d1351b&is=65bec01b&hm=1ae2c706edebd2817c26dc05c630b1eea956153c15463d923af1ff4608681a6b&=&format=webp&quality=lossless&width=1390&height=1176

BadgerWi Johnson

It worked thank you! Now I have new problem the .json file that I created in Kohay is not being seen in stable diffusion when I go to look at and load my loras. Loras I got from you or others I grabbed from CIVITAI show up just not he one I created in Kohay.

FREE AI CREATOR

Aitrepreneur Genius! But I still have a general question. I would like to ask for Styles there are cool styles such as "Snow" or "Mars" Which were for InstantID were on huggingface. hint Who knows how it is possible to install them in SD to create masterpieces, according to the lessons of Aitrepreneur. )))) ?)

Aitrepreneur

you can go inside the style_template.py file in the INSTID folder and inside you will see all the styles and prompts for those styles

Fatty McFatass

would you consider doing a video and script on SAD talker at some point please :)

Cisco Q

I'm not seeing the stable diffusion folder within the text generation ui folder. Can you direct me to the right place?

Keiron Gulrajani

is there a google colab set up for A1111 by any chance?

Aitrepreneur

You can't use a1111 with colab anymore, google started blocking sd webui a few month ago, you can use runpod instead but you will need to pay for it

Keiron Gulrajani

Thanks for the clarification, I'll check out runpod

Kythe Druid

Can you do a video about the sadtalker extension?

Silverback

I am on a apple device and sometimes get the errormessage, that anec....is not within range. And my other browser input then laggs. Anybody else got this problem?

Ron Harris-White II

Another question, is it possible to make InstantID public, as in 0.0.0.0 so I can access it on the local network vs public network via gradio?

Ron Harris-White II

If anything, as there be a thought about creating a pure AI workstation where you have everything (A1111, ComfyUI, Fooocus, InstantID, Ooogabooga, SillyTavern) installed and accessible via LAN (and let the user expose it to the internet if they are so inclined)?

Aitrepreneur

yes you need to modify the launcher and the --share argument

Aitrepreneur

you can use multiple runpod templates for that

Sid (edited)

Comment edits

2026-09-08 20:00 stable cascade coming soon i assume?
2024-02-14 04:03

Kamuro (edited)

Comment edits

2026-09-08 20:00 hi, I ran into such a problem, after the first launch, image generation on the rtx3080 was quite fast, less than 1 minute. after several generations, the speed has dropped significantly and now it takes from 5 minutes or more to generate one image. enabling low vram does not help. how can the problem be solved?
2024-02-15 17:23

Aitrepreneur

first make sure to dl the latest nvidia drivers, edit the webui-user.bat file and add the --xformers argument on the set command line args = if that doesn't make it better try --opt-sdp-attention instead

andrey lineyka

Thanks for the advice, I'll do as you said.

Xom

Hi, the installation failed , "windows can t find "webui-user.bat" I ran the installer in the stablediffusion-webui folder, and i can see the webui-user.bat in the folder. Any toughts ? Thanks :)

Xom

Ok my Ui was outdated, it s ok now :)

Morgan

I get an odd error message when I try to generate an image. AttributeError: module 'torch.nn.functional' has no attribute 'scaled_dot_product_attention'

Christopher Leblanc

Having some trouble following the video for some reason. It's super quick. 1st though... when I run the 1 click installer, I get some errors at the very end of it. stderr: WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'D:\\AI-Stable Diffusion\\stable-diffusion-webui\\venv\\Lib\\site-packages\\~afetensors\\_safetensors_rust.cp310-win_amd64.pyd' Check the permissions. WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) WARNING: Ignoring invalid distribution -rotobuf (d:\ai-stable diffusion\stable-diffusion-webui\venv\lib\site-packages) [notice] A new release of pip available: 22.2.1 -> 24.0 [notice] To update, run: D:\AI-Stable Diffusion\stable-diffusion-webui\venv\Scripts\python.exe -m pip install --upgrade pip Press any key to continue . . . D:\AI-Stable Diffusion\stable-diffusion-webui> 2nd... your video shows you uploading the pics, but you launch python and move back and forth superfast. Even slowing you down, I can't tell what you're doing. Can you explain a little bit why you launch python and what you're running there to get to the uploader?

lone wolf

there is a error , as disconnected when ever i try to use it

lone wolf

2024-03-09 19:57:28,259 - ControlNet - INFO - Loading model: control_instant_id_sdxl [c5c25a50] 2024-03-09 19:57:28,466 - ControlNet - INFO - Loaded state_dict from [D:\stable diffision AI\stable-diffusion-webui\models\ControlNet\control_instant_id_sdxl.safetensors] 2024-03-09 19:57:28,476 - ControlNet - INFO - controlnet_sdxl_config 2024-03-09 19:57:41,122 - ControlNet - INFO - ControlNet model control_instant_id_sdxl [c5c25a50](ControlModelType.InstantID) loaded. 2024-03-09 19:57:41,402 - ControlNet - INFO - Using preprocessor: instant_id_face_keypoints 2024-03-09 19:57:41,403 - ControlNet - INFO - preprocessor resolution = 512 2024-03-09 19:57:42,622 - ControlNet - INFO - ControlNet Hooked - Time = 24.702264070510864 100%|██████████████████████████████████████████████████████████████████████████████████| 10/10 [00:08<00:00, 1.13it/s] Total progress: 100%|██████████████████████████████████████████████████████████████████| 10/10 [00:04<00:00, 2.37it/s]

lone wolf

when ever i use the control net i get error , Error Connection errored out. but the cmd is connected

Twisted Tiger

yeah this thing isnt working for me, I will try through VPN sometimes that works

Twisted Tiger

this helped me, it is very important info, the thing is now looking a little different, hope it works this time

Twisted Tiger

I keep getting error 2. it just flat out wont work with the knowleged I have, please help.

lone wolf

yes we are getting error, the issue i foud that the models are not there in the model folder, and even we are dowloading the models, its not fetching it automatic, we are not able to use the basic open pose, how can we get the basic open pose back as , also inpating tab we are not able to use, can you help to reset

Aitrepreneur

I need more info, dm me the full error log, maybe screenshots also

Sebastian Paulmann

Amazing! Can i use this in Forge too? Manual install gave me trouble there before

Aitrepreneur

I'm not sure I haven't tried, might not be compatible but try it out, can't hurt to try :)

Sebastian Paulmann

Didnt really work on Forge but i just installed a second A111 instance, thank you! Forge is great though, this is the only downside for me

Jonathan Moreno

no instantid model in the drop down menu after install

Aitrepreneur

have you relaunched the webui after install? was there any error during installation?

Caio Vinicius

Maybe Launcherv2 isn't work anymore:

Caio Vinicius

D:\InstantID\INSTID>CALL env\Scripts\activate D:\InstantID\INSTID\env\Lib\site-packages\transformers\utils\generic.py:441: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead. _torch_pytree._register_pytree_node( D:\InstantID\INSTID\env\Lib\site-packages\transformers\utils\generic.py:309: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead. _torch_pytree._register_pytree_node( D:\InstantID\INSTID\env\Lib\site-packages\diffusers\utils\outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead. torch.utils._pytree._register_pytree_node( D:\InstantID\INSTID\env\Lib\site-packages\diffusers\utils\outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead. torch.utils._pytree._register_pytree_node( INFO:httpx:HTTP Request: GET https://api.gradio.app/gradio-messaging/en "HTTP/1.1 200 OK" Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: ./models\antelopev2\1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: ./models\antelopev2\2d106det.onnx landmark_2d_106 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: ./models\antelopev2\genderage.onnx genderage ['None', 3, 96, 96] 0.0 1.0 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: ./models\antelopev2\glintr100.onnx recognition ['None', 3, 112, 112] 127.5 127.5 Applied providers: ['CPUExecutionProvider'], with options: {'CPUExecutionProvider': {}} find model: ./models\antelopev2\scrfd_10g_bnkps.onnx detection [1, 3, '?', '?'] 127.5 128.0 set det-size: (640, 640) INFO:httpx:HTTP Request: GET https://checkip.amazonaws.com/ "HTTP/1.1 200 " INFO:httpx:HTTP Request: GET https://api.gradio.app/pkg-version "HTTP/1.1 200 OK" Current resolution settings: max_side = 1280, min_side = 1024 Arguments currently in use: inbrowser: True vae\diffusion_pytorch_model.safetensors not found Keyword arguments {'safety_checker': None} are not expected by StableDiffusionXLInstantIDPipeline and will be ignored. Loading pipeline components...: 86%|██████████████████████████████████████████ | 6/7 [00:18<00:03, 3.71s/steps]D:\InstantID\INSTID\env\Lib\site-packages\diffusers\utils\outputs.py:63: UserWarning: torch.utils._pytree._register_pytree_node is deprecated. Please use torch.utils._pytree.register_pytree_node instead. torch.utils._pytree._register_pytree_node( Loading pipeline components...: 100%|█████████████████████████████████████████████████| 7/7 [00:18<00:00, 2.61s/steps] Pressione qualquer tecla para continuar. . . After load everything... screen crashes

Shaun Newton

AutoInstaller and Manual Install doesn't work with controlnet 1.1.455 - wont show InstantID tab in Control type.

Shaun Newton

Demo also doesn't work on collab.

Aitrepreneur

you need to rollback to a previous controlnet version, the latest update breaks the instantid extension