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 HIDREAM TEXT-TO-IMAGE NEW AI KING ComfyUI WebUI! (Patreon)

Published:
2025-05-05 18:37:18
Edited:
2025-05-06 14:05:48
Imported:
2026-06
Tags:

Downloads

Content

EDIT: TO USE THE WORKFLOW DO NOT FORGET TO UPDATE COMFYUI IF YOU ARE USING AN OLDER VERSION OF COMFYUI! Simply updating comfyui to the latest version is enough if you have used the HIDREAM-MODELS-NODES_INSTALL.bat installer. DO NOT use the manager to update comfyui, the manager has been broken for months, to update comfyui correctly, go inside the update folder and run the update_comfyui_and_python_dependencies.bat file (do not forget to continue the install once it asks you to press any button) then, once the update is done, relaunch comfyui and re drag and drop the workflow inside.Hey everyone! I've created a 1-click installer for the newest "NEW IMAGE GEN KING ! ", called "HIDREAM". Simply THE BEST AND MOST POWERFUL AI TEXT TO IMAGE AI MODEL TO DATE! It's just amazing!You can check out the video right here: https://youtu.be/OeKRc2CjAmAThe workflow: https://www.patreon.com/posts/hidream-easy-to-128219569The 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!!!1. Download the HIDREAM-COMFYUI-MANAGER_AUTO_INSTALL.bat.bat

2. Run the bat file and select either option 1, 2 or 3 depending on your GPU vram and then Y or N if you want to install the FP8 Model

3. Wait for the install to finish

4. The webui will launch, load the HIDREAM_EASY-WORKFLOW.json file and there you go😎

IF YOU ALREADY HAVE COMFYUI INSTALLED USING A PREVIOUS INSTALLER:

1. Download the HIDREAM-MODELS-NODES_INSTALL.bat and place it in the ComfyUI_windows_portable\ComfyUI folder

2. Run the bat file and select either option 1, 2 or 3 depending on your GPU vram and then Y or N if you want to install the FP8 Model

3. Wait for the install to finish

4. ???

5. 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, 4090 or an A30 are great), click here for the comfyui template (aitrepreneur/comfyui), 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 HTTP SERVICE port 8888"

5. Then go inside the comfyui folder and then drag and drop the HIDREAM_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 HIDREAM_AUTO_INSTALL-RUNPOD.sh./HIDREAM_AUTO_INSTALL-RUNPOD.sh7. Press Y to dl the Fp8 model and wait for everything to be installed

8. Go back to the port 3000 and click on the manager to update comfyui and restart it

9. Load the HIDREAM_AUTO_INSTALL-RUNPOD.sh 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:1. Go inside your pod2. SERVICE port 88883. Go inside the Comfyui folder, open a terminal icon and re copy and paste those two lines:chmod +x HIDREAM_AUTO_INSTALL-RUNPOD.sh./HIDREAM_AUTO_INSTALL-RUNPOD.sh4. Press Y and then in a few seconds it will reinstall all the requirements for the nodes5. Go back to comfyui port 3000, click on manager, click update all, wait for everything to be updated then restart.6. Re drag and drop the HIDREAM_AUTO_INSTALL-RUNPOD.sh 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😉!ps: Sorry for the lack of news lately, I've already talked about it before, so no need to talk about it again but thanks for your support and patience, much love❤️

Files

Comments

bwild370

I'm new to AI and always have a lot of questions haha. The first one is where is the HIDREAM_EASY-WORKFLOW.json file. I'm blind sometimes haha.

rubz

In the other post, not this one

ShmentaixLord

how much does it weight with FP8s included?

Colby´s HangBeats

Hi, you have presented a website that presents the different artists. Can you please provide the link. I can't find this page. Thanks.

Audiogus

-------- Checking prerequisites -------- 7ΓÇæZip not found ΓÇô downloading... % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 154 100 154 0 0 172 0 --:--:-- --:--:-- --:--:-- 172 0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0 100 1538k 100 1538k 0 0 926k 0 0:00:01 0:00:01 --:--:-- 926k 7ΓÇæZip install failed ΓÇô install it manually then rerun this script. Press any key to continue . . .

bwild370

Thank you lol. I always miss it lmao.

Audiogus

So looks like I had to add "chcp 65001" to the beginning of the batch file to parse the text properly, do a find and replace on dashes "-" to be "-" (?) and add my 7zip to the windows PATH and now the batch file is working. Also i had to run it in Powershell and not a regular DOS prompt.

JeanMarc

I have a super anime/fake renders even with tailored pos / neg prompts. Am I missing something?

Nibmeister

Thanks for this. It works well on my 3070TI with 8 GB VRAM.

Midnight

I get this error when using the workflow. Unexpected architecture type in GGUF file, expected one of flux, sd1, sdxl, t5encoder but got 'hidream' I used the update installer that puts in new nodes and and models. I selected the GGUF 5 in the GGUF loader. I get the above error message. Any ideas? Thank you.

Kusacubari

Even when i clean load the 1-click install - I still end up with a "missing node" error for WORKFLOWLOADER where can I get it please

RJ Nash

Same... I'm unsubbing.. never had any luck with his BAT files.. better use Docker or Pinokio or any other APP to get these toys working easily within an ENV.

Aitrepreneur

every single models combined is around 40gb

Aitrepreneur

depends on your prompt I suppose, you can take inspiration of a few prompts on civitai for example

Aitrepreneur

why haven't you send me a dm? I can't see all the messages I get, hence why I'm asking people to dm me their issues instead?

Aitrepreneur

if you are using an already existing comfyui version, once you have updated comfyui to the latest version and all the nodes, just redrag and drop the workflow inside comfyui again and select the corresponding GGUF model for your GPU (in your case it will be the Q5 ks)

Aitrepreneur

if you are using an already existing comfyui version, once you have updated comfyui to the latest version and all the nodes, just redrag and drop the workflow inside comfyui again

Aitrepreneur

Send me a dm this is not normal, this is literally working for me rn, it's the easiest stuff ever on runpod

Aitrepreneur

Send me a dm this is not normal, this is literally working for me rn, it's the easiest stuff ever on runpod

XJollyboyx

mine fails at extracting comfyui

clearandsweet

Says incompatable with 64 bit windows on the installer

clearandsweet

Figured it out by plugging it into Gemini 2.5 lololo didn't have Git installed

Kusacubari

i did try using my original comfyui but its didn't work out well, so I went with a clean run with the 1-click and redownloaded the models to be sure and clean but still the worflowloader" node is red. I can see the text inside it says "pasted/image (15).png" so its not missing. its just red and dead. Im about to try replacing it with a standard image loader instead

Midnight

Okay so this fix did not work. But reinstalling from scratch with your new bat file seems to have fixed it. No issue at all. It was just not playing well with updating from existing install. Thank you.

Kusacubari

but did get to try the hidream flows that did work and its v impressive for sure, nice one

Andrew Van

what to do about 64-bit windows incompatibility error?

Vanz Watts

i get to 20% and it crashes. not sure what i am doing wrong. fresh install.

Peter L

Any idea why whenever i run one of these BAT files it just downloads into my system32 folder instead of where i launch it from?

Aitrepreneur

Because you are running it as admin, just run it normally

Peter L

okay let me try that, thanks

Aitrepreneur

Dm me error logs or screenshots or video recording whatever you ca

Darren Keathley

for some reason, not that it is a big deal, I cannot get any NSFW stuff, like even if I go full hard core Phub style request it will just give me 2 girls in a bikini

Yu Yeh

This is great. Would yoi be able to create a tutorial for training a LoRa using HiDream ?

JeanMarc

pretty generic "photorealistic, beautiful french woman with thin wavy short dark hair with bangs and blue eyes wearing a white dress with a red floral pattern and a beret girl with cute french outfit and a french beret with streets of paris in the background, realistic skin texture, natural light, shallow depth of field, 85mm lens, Canon EOS 5D, photorealistic, unedited photo" my negative are "bad proportions, ugly, bad art, bad anatomy, negative_hand-neg,poorly drawn hands, poorly drawn face, fused fingers,too many fingers, not enough fingers, bad, blurry, anime, cartoon, unrealistic, over-smoothed, doll-like"

Kusacubari

Still get the "workflowloader" node in red even after clean 1-click install and update comfy in manager and manually. That node still failing

James Bowler

me too, I tried the install nodes.bat did the update and tried dragging the workflow back in and I get Missing Node types>ImageResizeKJv2 (in groupnode 'workflow>loader', the remove from workflow. UltimateSDUpscale workflowloader, I go into manager to install missing nodes and I see: IMPORT FAILED ↗ ComfyUI_UltimateSDUpscale Try fix, Switch, Unistall, etc...

Todd Keck

I can't get the workflowloader node to work either

LW

Just watched your HiDream video and..... WOW, AI is moving fast. Good job we have K to keep us... er.. abreast of the latest developments and make installation simple.

Aitrepreneur

I guess so but it will all depends on the community and what model they chose to use

Aitrepreneur

Simply updating comfyui to the latest version is enough if you have used the installer. DO NOT use the manager to update comfyui, the manager has been broken for months, to update comfyui correctly, go inside the update folder and run the update_comfyui_and_python_dependencies.bat file (do not forget to continue the install once it asks you to press any button) then, once the update is done, relaunch comfyui and re drag and drop the workflow inside.

Aitrepreneur

Simply updating comfyui to the latest version is enough if you have used the installer. DO NOT use the manager to update comfyui, the manager has been broken for months, to update comfyui correctly, go inside the update folder and run the update_comfyui_and_python_dependencies.bat file (do not forget to continue the install once it asks you to press any button) then, once the update is done, relaunch comfyui and re drag and drop the workflow inside.

Onex Sys

I randomly get flowers every now and then in my results - be it as part of the image or weird overlays. There is absolutely nothing in my prompts that would refer to anything floral. Did I do something wrong settings wise?

Aitrepreneur

probably because you activated the add more details option at the bottom, sometimes the effect can be kinda random, you can either disable the option or decrease the value (read the note next to the node)

Aitrepreneur

Simply updating comfyui to the latest version is enough if you have used the installer. DO NOT use the manager to update comfyui, the manager has been broken for months, to update comfyui correctly, go inside the update folder and run the update_comfyui_and_python_dependencies.bat file (do not forget to continue the install once it asks you to press any button) then, once the update is done, relaunch comfyui and re drag and drop the workflow inside.

Onex Sys

I already suspected that, but it still occurs :(

Aitrepreneur

DM me your prompt and screenshots of the parameters

Kusacubari

as @Aitrepreneur said to me: use the update folder and run the update_comfyui_and_python_dependencies.bat and then the update_comfyui.bat which fixed the node thanks man

GlobalGrooveWORLD

I did everything as you said, updated comfy via bat file. In the new manager I clicked update missing nodes, closed comfy in general and reloaded it again, closed workflow and threw the script again and still this node is not there! There are no missing nodes in the manager as well. Here is what version I have ComfyUI Manager V3.31.1. And how to fix this problem?

studiosmh

Hidream works well, thanks for your great work. I have been trying to get Reactor face swapper to load (the uncensored one not the SFW version) but it fails to load in my new comfyui with hidream. Any chance you will do an installer for it and add Reactor to your Hidream workflow?

Aitrepreneur

I have the face swap in my ultimate workflow V3, I might have to segment it into smaller workflows for people to use though but it should work already with the current installer.

Aitrepreneur

You don't need to update missing node, have you used the correct HIDREAM-MODELS-NODES_INSTALL.bat file inside the comfyui folder?

studiosmh

That would be great thanks.

Lars Drakblod

There was some mentioning of this model being uncensored... *cough* my friend *couch* said only SFW picture got produced... *cough* any parameters of files that need to be change for more spicy picture... my friend wonders? ;-D

Nomeora

Okay that's gotta be my issue as well. You should mention that in the video. A lot of people looking for 1 click installers are probably also noobs like me!

phail216

Yeah, compared to SDXL this is censored af

Aitrepreneur

I mean...you are probably talking about a CUSTOM TRAINED sdxl model not a base model, custom models are trained specifically on hardcore images...have you ever tried any base model to make the same fair comparaison?

Lars Drakblod

were does one find Loras that are compatible with HiDream?

GlobalGrooveWORLD

ImageResizeKJv2 (In group node 'workflow>loader')

Aitrepreneur

yeah as I explained, you need to correctly updated comfyui using the update bat file, your comfyui needs to be at version 0.3.31 (2025-05-03) scroll down the manager, look here: https://imgur.com/OOaldHr then once it's updated just redrag and drop the workflow back in again

GlobalGrooveWORLD

The link doesn't work, but if I got it right myself, here's what it says at the very bottom ComfyUI: v0.3.32 (2025-05-06) Manager: V3.31.13 So I have a newer version and need to look for an older version?

Aitrepreneur

Nope that's great, you got the very last version, which is another problem then...let's go in dm then

Tyme Cich

If you find a resolution for this, please share. I'm having the same issue. ComfyUI: v0.3.32-2-g271c9c5b (2025-05-06) Manager: V3.31.13

Groot Sanchez

ngl I love working on runpod.io, but if you can accept some constructive criticism. You gotta improve your instructions a little bit better. I was finally able to fully up and run runpod.io with your template properly after some tinkering. I'm a software engineer and I'm super thorough with instructions. Let me know if I can help.

Aitrepreneur

Thanks but I mean I litterally show everything step by step in my video, can't be more visual and clear than that

GlobalGrooveWORLD

It might have to do with your VPN as well. First, go to the, comfyui/custom nodes folder and delete the 8 folders: ComfyUl_essentials ComfyUl_UItimateSDUpscale ComfyUl-Detail-Daemon ComfyUl-GGUF ComfyUl-KJNodes ComfyUl-Manager rgthree-comfy wlsh nodes then disable your vpn, and relaunch the nodes/models installer in the comfyui folder. then relaunch comfyui and redrag and drop the workflow again inside

Kirk Cowgill

I did everything and then finally dragged the workflow into comfyui. Nothing appears. There's a tab at the top showing the name of the workflow, but the grey-black field is completely blank and empty.

Darkhitman66

dang 41 min to generate 3060 12 gb ram the q4 model

Darkhitman66

ok 2nd time doing the gen it did it in 9 min so first gen takes a long time lol

Groot Sanchez

I forgot you made videos haha. That's how I found your patreon

X.Codex

Great work with the installers. Ran into an issue because my comfyui was outdated with the node installer so decided to run the auto_install, and it worked perfect. That could not of been easier.

LW

Updating ComfyUI was definitely necessary for me (K, thanks for the tip about using update_comfyui_and_python_dependencies.bat rather than the Manager). Installing HiDream in to my existing ComfyUI directory with HIDREAM-MODELS-NODES_INSTALL.bat didn't work for me. It installed files in to the custom-nodes folder. However, doing a full fresh install with HIDREAM-COMFYUI-MANAGER_AUTO_INSTALL.bat worked fine. HiDream is impressive, great realism, fantastic hands.... five digits per hand... amazing where previous models got so many images of people with 6 fingers, polydactyly, I believe! The HiDream guys must have used less images of Siamese Twins too as I get less two headed people or people with extra limbs. The image quality is great. What's needed now is Loras for all those special situations that probably weren't included in the data set ;) Has FluxGym been updated for HiDream, or is there going to be a HiDream Gym release?

Garion128

I updated ComfyUI as you said, but since I can't run "Run nvidia Gpu.bat" I get the AssertionError: Torch not compiled with CUDA enabled I tried forcing reintalling pytorch with cuda enabled with this : "pip install --force-reinstall torch --index-url https://download.pytorch.org/whl/cu128" But it doesn't worked at all... Do I need to make a complete fresh reinstall, or is there a way to save mine ? Thanks !

Randy Monteith

I am trying to figure out this "workflowloader" what is it ? Its still red in two sections of the HiDream workflow

Moses Son

how do we train loras?

Moses Son

do you know if civitai supports training hidream loras?

Moses Son

solution: was getting the error "Unexpected architecture type in GGUF file, expected one of flux, sd1, sdxl, t5encoder but got 'hidream'", i fixed this by going in the manager and looking for node updates, nothing, i searched the node "gguf" from the list of up to date nodes and forced the "gguf" node to update, restarted the manager an BOOM its working now.

Gabo

same for me

Aitrepreneur

click the fit the view icon on the bottom right

BTS

fresh Install via one-click. All I get is a black empty image.?? Is rgThree-comfy supposed to be in the node structure? TIA for anyone's help.

Sean Rowe

I can't run the auto installer for ComfyUI as I'm getting an error: Unsupported 16-bit application: The program or feauture "filepath/git-setup.exe" cannot start or run due to incompatibility with 64-bit versions of Windows. Please contact software vendor to aks if a 64-bit Windows compatible version is avaiable. I have ComfyUI installed but I did it through StabilityMatrix I'm wondering if it's fully updated. I can't run the update_comfyui_and_python_dependencies.bat on that since it builds it's own git environment. But I'm ok with moving to a stand-along ComfyUI however as mentioned the one click installer doesn't work for me.

Aitrepreneur

you need to install git for windows manually

Neil H.

Thank you so much! It is working well on my end. Is it possible to use IPadapter with HiDream?

Russ Ader

ERROR: Exception: Traceback (most recent call last): File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\urllib3\response.py", line 438, in _error_catcher yield File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\urllib3\response.py", line 561, in read data = self._fp_read(amt) if not fp_closed else b"" ^^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\urllib3\response.py", line 527, in _fp_read return self._fp.read(amt) if amt is not None else self._fp.read() ^^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\cachecontrol\filewrapper.py", line 98, in read data: bytes = self.__fp.read(amt) ^^^^^^^^^^^^^^^^^^^ File "http\client.py", line 479, in read File "socket.py", line 720, in readinto File "ssl.py", line 1251, in recv_into File "ssl.py", line 1103, in read TimeoutError: The read operation timed out During handling of the above exception, another exception occurred: Traceback (most recent call last): File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\cli\base_command.py", line 106, in _run_wrapper status = _inner_run() ^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\cli\base_command.py", line 97, in _inner_run return self.run(options, args) ^^^^^^^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\cli\req_command.py", line 67, in wrapper return func(self, options, args) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\commands\install.py", line 386, in run requirement_set = resolver.resolve( ^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\resolution\resolvelib\resolver.py", line 179, in resolve self.factory.preparer.prepare_linked_requirements_more(reqs) File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\operations\prepare.py", line 554, in prepare_linked_requirements_more self._complete_partial_requirements( File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\operations\prepare.py", line 469, in _complete_partial_requirements for link, (filepath, _) in batch_download: ^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\network\download.py", line 184, in __call__ for chunk in chunks: ^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\cli\progress_bars.py", line 55, in _rich_progress_bar for chunk in iterable: ^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_internal\network\utils.py", line 65, in response_chunks for chunk in response.raw.stream( ^^^^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\urllib3\response.py", line 622, in stream data = self.read(amt=amt, decode_content=decode_content) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\urllib3\response.py", line 560, in read with self._error_catcher(): ^^^^^^^^^^^^^^^^^^^^^ File "contextlib.py", line 158, in __exit__ File "D:\050725Hidream\ComfyUI_windows_portable\python_embeded\Lib\site-packages\pip\_vendor\urllib3\response.py", line 443, in _error_catcher raise ReadTimeoutError(self._pool, None, "Read timed out.") pip._vendor.urllib3.exceptions.ReadTimeoutError: HTTPSConnectionPool(host='files.pythonhosted.org', port=443): Read timed out.

Kirk Cowgill

That wasn't it. I zoomed out infinitely to complete emptiness. The problem was that the auto-installer missed several items. I went through your manual instructions and found missing files. After downloading them manually and putting them into the right folders, it worked. I suggest that you have a way for the auto-installer to be run again without it forcing everything to start over from the beginning. Like, a 2nd launch of it would find the missing parts and put those in instead of starting from scratch.

deadwolf1227

been trying to get this to work for days. any idea how to fix a error UnetLoaderGGUF "cannot reshape array of size 1588364 into shape (2560,4752)"? Its a fresh install using the auto installer, i have the right model selected and updated every node.

Bill P

is there any way to change the script to contioune after it errors out and not restart the file each time? i have been trying to download 2 day now

Bill P

-------- 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 43 1974M 43 850M 0 0 355k 0 1:34:41 0:40:45 0:53:56 421k curl: (18) end of response with 1179463021 bytes missing Download failed. Press any key to continue . . . Terminate batch job (Y/N)?

Vincent Blokker

I'm getting install errors: curl: (23) Failure writing output to destination, passed 16384 returned 0 [ RealESRGAN_x4plus_anime_6B.pth ÔÇó downloading ae.safetensors % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 100 1106 100 1106 0 0 5176 0 --:--:-- --:--:-- --:--:-- 5241 0 319M 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 curl: (23) Failure writing output to destination, passed 15611 returned 0 [ ae.safetensors ------------------------------------------------------------- Install complete ÔÇô launching ComfyUI now ------------------------------------------------------------- 'run_nvidia_gpu.bat' is not recognized as an internal or external command, operable program or batch file. Press any key to continue . . .

LW

Just did a quick scan of the internet for HiDream Lora training and.....OMG.... K! K! We need your help!

Shokku

Newly build pc and fresh install - I installed Git and used your all-in-one installer, been nothing but errors and red text. I have been using A1111 SD since from the beginning and very first time trying to use ComfyUI. Have I missed something? Your workflow is not also loading for me when I drag and drop it.

Aitrepreneur

you need a specific hidream ipadapter for that, which doesn't exist yet

Aitrepreneur

this is usually because of a corrupted model file, delete the model in question and use the models node installer to reinstall it

Aitrepreneur

seems like an internet connection error issue, or a vpn. Disable your vpn (if you have one) and try again not much you can do for that. Could also have been an issue with huggingface connection, try again

Aitrepreneur

seems like an internet connection error issue, or a vpn. Disable your vpn (if you have one) and try again not much you can do for that. Could also have been an issue with huggingface connection, try again.

Aitrepreneur

need more info here, dm me some screenshots and the error log, for the "empty" workflow, just click the icon on the bottom right to fit into view

Martin Thoburn

HIDREAM-MODELS-NODES_INSTALL ends up with missing nodes when workflow is opened. Comfy Manger won't find missing nodes either.

Martin Thoburn

Derp, updating comfy and python fixed the issue

Jonathan Jewett

Try using the Fit View button in the bottom right corner. That's what did it for me. The workflow wasn't centered in the screen.

LW

Fantastic! Thanks, K :)

Randy Monteith

I did Update KJNodes still got the error but at first I did not reload the fresh workflow and after reading your message I did reload the workflow and not it works! Thanks for getting back to me!

Neil H.

Ah, ok. Thank you for the information!

Shokku

I kinda got it working. Installed the basic ComfyUI, drag and dropped the HIDREAM files into the ComfyUI folder replacing everything. Now I am only getting an error when trying to run the update_comfyui_and_python_dependencies. https://imgur.com/DGyOOCc Also, thanks, Jonathan. I would be grateful if Aitrepreneur could make a basic tutorial video how to use ComfyUI in 2025, such as using nodes, updating, generating images or videos, etc.

The Capn

Lesson learned... don't install by running the .bat as administrator... installs to System32, C drive weeps in 30gb free space...

Christopher Carlson

The pictures are clear, but at 1024 x 1024 I am getting weird tile often times around the edges. Has anyone experienced this? What is the best image size?

Chris Angelo

Correct. Move the comfyui-portable folder and you should be good.

tramidus mondge

hello, when i execute hidream-confyui-manager-auto-install.bat i have an error can you help me plz ? : 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. Error: Failed to install SD Scripts dependencies. Appuyez sur une touche pour continuer...

Micheal Hernandez

Curious about lora use, haven't had any luck on my own. Seems like flux loras may not be cross architecture Compatible. Any ideas?

Aitrepreneur

yeah loras are not cross architecture compatible as you said, no matter the model hence why you need specific lora trained on the model you want to use

LW

Fear not Micheal, our tireless AI Overlord K is on the job. (He told me so :) )

Aitrepreneur

yeah except I'm hesitating doing a full video on it actually, you train a lora for hidream, even a 4090 is almost not enough for the training, so it's either runpod or wait until things get more optimized but even then most people without a big 24gb vram gpu wouldn't be able to train a lora for hidream...

Nicolas Giarrusso

Missing Node Types When loading the graph, the following node types were not found ImageResizeKJv2 (In group node 'workflow>loader') missing nodes manager dont find the node to install it

ChudJones

I saw the same thing and wondered why no one mentions it. It looks really bad. At 1024x1024, I keep Shift at 2.00 or less and the artifacts don't show up as much.

Aitrepreneur

click on the manager and click update all, to update all the nodes, then re drag and drop the workflow inside comfyui again

LW

Indeed, I read that you need a 24gb gpu (or better). Perhaps some people would use Runpod. I do have a 24gb gpu and don't mind if it takes some time... in the end, I can always leave it crunching the pixels overnight. The question I have is, which software to use? The videos you have done for making loras for other models have been very helpful and certainly give me a head start, so anything you can do, even a quickie, would be welcome. :)

Soham Dasgupta

Can you please share the reference website where they show all the styles of photos HiDream can produce?

Aitrepreneur

I will do a video anyway I think, I really want this model to work so I'll promote it any way I can until we have something better of course.

kirazo

It show cannot start or run with 64bit version of windows, what should I do? (Comfymanager)

aero costaa

its installed but it doesnt open the workflow, when i manually drag and drop it, it fails to open it

aero costaa

Prompt execution failed Prompt has no outputs:

aero costaa

# ComfyUI Error Report ## Error Details - **Node ID:** N/A - **Node Type:** N/A - **Exception Type:** Prompt execution failed - **Exception Message:** Prompt has no outputs: ## Stack Trace ``` Error: Prompt execution failed at ComfyApi.queuePrompt (http://127.0.0.1:8188/assets/index-CJpGM-n7.js:96128:13) at async PromptService.api.queuePrompt (http://127.0.0.1:8188/rgthree/common/prompt_service.js:88:28) at async ComfyApp.queuePrompt (http://127.0.0.1:8188/assets/index-CJpGM-n7.js:257267:25) at async app.queuePrompt (http://127.0.0.1:8188/extensions/rgthree-comfy/rgthree.js:463:24) at async function (http://127.0.0.1:8188/assets/GraphView-Ce_Nq6v7.js:4966:9) at async http://127.0.0.1:8188/assets/index-CJpGM-n7.js:190382:14 at async Proxy.execute (http://127.0.0.1:8188/assets/index-CJpGM-n7.js:192069:7) at async queuePrompt (http://127.0.0.1:8188/assets/GraphView-Ce_Nq6v7.js:4179:7) ``` ## System Information - **ComfyUI Version:** 0.3.30 - **Arguments:** ComfyUI\main.py --windows-standalone-build - **OS:** nt - **Python Version:** 3.12.9 (tags/v3.12.9:fdb8142, Feb 4 2025, 15:27:58) [MSC v.1942 64 bit (AMD64)] - **Embedded Python:** true - **PyTorch Version:** 2.7.0+cu128 ## Devices - **Name:** cuda:0 NVIDIA GeForce RTX 4070 Laptop GPU : cudaMallocAsync - **Type:** cuda - **VRAM Total:** 8585216000 - **VRAM Free:** 7398752256 - **Torch VRAM Total:** 0 - **Torch VRAM Free:** 0 ## Logs ``` 2025-05-18T05:57:55.880423 - [START] Security scan2025-05-18T05:57:55.880423 - 2025-05-18T05:57:56.596697 - [DONE] Security scan2025-05-18T05:57:56.596697 - 2025-05-18T05:57:56.696102 - ## ComfyUI-Manager: installing dependencies done.2025-05-18T05:57:56.696102 - 2025-05-18T05:57:56.696102 - ** ComfyUI startup time:2025-05-18T05:57:56.696102 - 2025-05-18T05:57:56.696102 - 2025-05-18 05:57:56.6962025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - ** Platform:2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - Windows2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - ** Python version:2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - 3.12.9 (tags/v3.12.9:fdb8142, Feb 4 2025, 15:27:58) [MSC v.1942 64 bit (AMD64)]2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - ** Python executable:2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - C:\Users\User\Downloads\ComfyUI_windows_portable\python_embeded\python.exe2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - ** ComfyUI Path:2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - ** ComfyUI Base Folder Path:2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.697101 - ** User directory:2025-05-18T05:57:56.697101 - 2025-05-18T05:57:56.699108 - C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\user2025-05-18T05:57:56.699108 - 2025-05-18T05:57:56.699108 - ** ComfyUI-Manager config path:2025-05-18T05:57:56.699108 - 2025-05-18T05:57:56.699108 - C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\user\default\ComfyUI-Manager\config.ini2025-05-18T05:57:56.699108 - 2025-05-18T05:57:56.699108 - ** Log path:2025-05-18T05:57:56.699108 - 2025-05-18T05:57:56.699108 - C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\user\comfyui.log2025-05-18T05:57:56.699108 - 2025-05-18T05:58:12.934882 - [ComfyUI-Manager] 'numpy' dependency were fixed 2025-05-18T05:58:12.935883 - Prestartup times for custom nodes: 2025-05-18T05:58:12.935883 - 0.0 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\rgthree-comfy 2025-05-18T05:58:12.936881 - 17.8 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager 2025-05-18T05:58:12.936881 - 2025-05-18T05:58:19.679635 - Checkpoint files will always be loaded safely. 2025-05-18T05:58:19.876774 - Total VRAM 8188 MB, total RAM 65256 MB 2025-05-18T05:58:19.876774 - pytorch version: 2.7.0+cu128 2025-05-18T05:58:19.877771 - Set vram state to: NORMAL_VRAM 2025-05-18T05:58:19.877771 - Device: cuda:0 NVIDIA GeForce RTX 4070 Laptop GPU : cudaMallocAsync 2025-05-18T05:58:28.164767 - Using pytorch attention 2025-05-18T05:58:41.397956 - Python version: 3.12.9 (tags/v3.12.9:fdb8142, Feb 4 2025, 15:27:58) [MSC v.1942 64 bit (AMD64)] 2025-05-18T05:58:41.398887 - ComfyUI version: 0.3.30 2025-05-18T05:58:41.546531 - ComfyUI frontend version: 1.17.11 2025-05-18T05:58:41.552538 - [Prompt Server] web root: C:\Users\User\Downloads\ComfyUI_windows_portable\python_embeded\Lib\site-packages\comfyui_frontend_package\static 2025-05-18T05:58:46.140684 - ComfyUI-GGUF: Partial torch compile only, consider updating pytorch 2025-05-18T05:58:46.178276 - Total VRAM 8188 MB, total RAM 65256 MB 2025-05-18T05:58:46.179550 - pytorch version: 2.7.0+cu128 2025-05-18T05:58:46.181260 - Set vram state to: NORMAL_VRAM 2025-05-18T05:58:46.182266 - Device: cuda:0 NVIDIA GeForce RTX 4070 Laptop GPU : cudaMallocAsync 2025-05-18T05:58:46.313589 - ### Loading: ComfyUI-Manager (V3.32.2) 2025-05-18T05:58:46.314590 - [ComfyUI-Manager] network_mode: public 2025-05-18T05:58:46.395670 - ### ComfyUI Revision: 163 [a97f2f85] *DETACHED | Released on '2025-04-24' 2025-05-18T05:58:47.122223 - 2025-05-18T05:58:47.122223 - ?[92m[rgthree-comfy] Loaded 46 fantastic nodes. 🎉?[00m2025-05-18T05:58:47.122223 - 2025-05-18T05:58:47.122223 - 2025-05-18T05:58:47.186576 - Import times for custom nodes: 2025-05-18T05:58:47.187584 - 0.0 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\websocket_image_save.py 2025-05-18T05:58:47.187584 - 0.0 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_UltimateSDUpscale 2025-05-18T05:58:47.187584 - 0.1 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\wlsh_nodes 2025-05-18T05:58:47.188576 - 0.1 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-GGUF 2025-05-18T05:58:47.189584 - 0.1 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_essentials 2025-05-18T05:58:47.190584 - 0.1 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\rgthree-comfy 2025-05-18T05:58:47.190584 - 0.2 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-KJNodes 2025-05-18T05:58:47.191585 - 0.6 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Manager 2025-05-18T05:58:47.191585 - 0.9 seconds: C:\Users\User\Downloads\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-Detail-Daemon 2025-05-18T05:58:47.192576 - 2025-05-18T05:58:47.232665 - Starting server 2025-05-18T05:58:47.233763 - To see the GUI go to: http://127.0.0.1:8188 2025-05-18T05:58:48.092681 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/model-list.json 2025-05-18T05:58:48.323585 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/alter-list.json 2025-05-18T05:58:48.523063 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/github-stats.json 2025-05-18T05:58:48.687482 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/extension-node-map.json 2025-05-18T05:58:48.910238 - [ComfyUI-Manager] default cache updated: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json 2025-05-18T05:58:49.696015 - [ComfyUI-Manager] An error occurred while fetching 'https://api.comfy.org/nodes?page=1&limit=30&comfyui_version=v0.3.30&form_factor=git-windows': Expecting value: line 2 column 1 (char 1) 2025-05-18T05:58:49.696015 - Cannot connect to comfyregistry.2025-05-18T05:58:49.696015 - 2025-05-18T05:58:49.700584 - FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json2025-05-18T05:58:49.700584 - 2025-05-18T05:58:50.596898 - got prompt 2025-05-18T05:58:50.597890 - invalid prompt: {'type': 'prompt_no_outputs', 'message': 'Prompt has no outputs', 'details': '', 'extra_info': {}} ``` ## Attached Workflow Please make sure that workflow does not contain any sensitive information such as API keys or passwords. ``` Workflow too large. Please manually upload the workflow from local file system. ``` ## Additional Context (Please add any additional context or steps to reproduce the error here)

James Hossler

I got all the way through the install and it shut off because I do not have a Nvidia GPU. Is there any way to make this work with AMD?

Paul Mazurkiewicz

Everyone needs to learn how to do this in Comfy...always push the little icon that looks like a square with bits missing from each of the four sides. It will solve 99% of the problems people report.

Paul Mazurkiewicz

click the little square on the lower right that looks like a square with bits of the sidewalls missing on each side. That will solve your problem.

HOLO

after trying to generate the first image I got this error: Prompt execution failed Prompt outputs failed validation: UnetLoaderGGUF: - Value not in list: unet_name: 'hidream-i1-dev-Q8_0.gguf' not in ['hidream-i1-dev-Q5_K_S.gguf']

Fizzy

I'm pretty sure I have a sampler selected and the node is set to always Prompt outputs failed validation: SamplerCustomAdvanced: - Required input is missing: sampler This happens if I have the add details node off. if it's on then I get Unexpected architecture type in GGUF file: 'hidream' Even after running the update bat file and everything getting the same error :(

Yura

Can i use HiDream Image 2 Image in ComfyUI ?

Shaun Szymanski

i keep getting an error when trying to download. "The program or feature "\??D:git-setup.exe Cannot start or run due to incompatibility with 64-bit version of windows.

Bryan Crump

are we able to train models for this one as well?

John Solo

how does one add a checkpoint loader into this workflow without breaking it?

BadgerWi Johnson

I keep getting an error Prompt execution failed Prompt outputs failed validation: UnetLoaderGGUF: - Value not in list: unet_name: 'hidream-i1-dev-Q8_0.gguf' not in ['hidream-i1-dev-Q4_K_S.gguf']

amazingghost

yeah same this is busted - dontthink it downloads the models anyore or he didnt really check it out fully

Xian S. D. Dorsey

Says the QuadrupleCLIPLoader is missing and i cannot find it. Also missing the HI Dream model. Any suggestions?

Wicked Desires

I'm getting this error: Prompt outputs failed validation: ImageResizeKJv2: - Value not in list: device: '2' not in ['cpu', 'gpu']

Charles Lundy

I think there are steps that I don't know when installing. I just finished this PC build so nothing is set-up. I grabbed the Cuda 12.6 that I saw in other posts. hmmm, I don't think Patreon will allow me to provide the screen shot of what happens when I open the auto installer... it's not all in English. as if it can't read the file or translate the code properly. uses symbols amongst the words in the command box.

Grey Walker

This eats up all my memory but barely touches the gpu (3060). I was able to get one image to render after a bunch of fresh reinstalls, but the next one froze the computer. Am I missing something? I have 32G of RAM which gets used upwards of 90%.

Alexandre

You need to change the model to the one you have chosen YOURSELF at the start of the .bat file.

Yu Yeh

Hello, I learned from your Kontext Lora training how we can use AI Toolkit UI to train a Lora for HiDream. is it pretty much the same process as training a LoRa for Flux.dev? Thank you

Malik

Doing runpod install. On the step to update ComfyUI from manager I get a "Failed to update ComfyUI" error. What do I do?

Oscar Petoskey

Sorry to say, this model is totally broken! Had a beautifully working install some months ago, ran it a couple days ago, and encountered many errors. Ran update comfyUI with python batch on Windows, still broken. Reinstalled into existing install, reinstalled all models... still broken. Tried fresh install into new directory, still broken after updates.