Aitrepreneur
patreon
HIDREAM EASY TEXT-TO-IMAGE WORKFLOW!
🕑 Added 2025-05-05 18:36:50 +0000 UTCEDIT: TO USE THE WORKFLOW DO NOT FORGET TO UPDATE COMFYUI! 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 just released a new video on "THE 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/OeKRc2CjAmA
I also created a 1-click installer for the Webui for my Patreon supporters for Local AND Runpod, that you can find right here: 1-Click INSTALL HIDREAM TEXT-TO-IMAGE NEW AI KING ComfyUI WebUI! | Patreon
Here the link to the manual install guide: https://pastebin.com/hvU8w7y5
Down below you will also find the HIDREAM_EASY-WORKFLOW.json for free for you to use!
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❤️
Comments
Onyx N
i think this workflow really need an update, its censored
Starbuckk Serapis
This is producing censored images. Am I mssing a step?
Omar Abouzaid
Do you ever plan to do a HiDream LoRA training guide video? You mentioned some interesting points in your video such as the fact that the dev model is better and that the provided settings were no optimal for training and I'd love to hear some more details about that as well as to know which training client you prefer (ai-toolkit, diffusion pipe, etc.)
George Thompson
I figured this out for anybody else that gets this error. in the workflow device option switch it to cpu or gpu
Aitrepreneur
click on the fit to view icon on the bottom right
Raevn Eldaer
The workflow is just empty when I load it...
George Thompson
Hey I downloaded it and did a fresh install and it works great but the image to image is broken on my install. says the following: Prompt execution failed Prompt outputs failed validation: ImageResizeKJv2: - Value not in list: device: '2' not in ['cpu', 'gpu'] every run across this error before? any idea how to fix it? it's just effecting the image to image
Extreme Mojo
This is absolutly awesome. I would love to see a video about how to train LORAs with one trainer. I can't seem to get it to work.
Stefan Mihai Creepypasta
I have the same issue :(
Onex Sys
After installing the Hidream "Update" to a prior installation of Flux with using your 3.1 Workflow - Suddenly the 3.1 Workflow gives me OOM exceptions when creating from text. Image to Image still works fine. I tried a completely new installation into a new folder of the "old" 3.1 workflow, but It leaves me with missing nodes (tried all your fixes from troubleshoot video) but still have those errors: workflowFlorence workflowFlux Controlnet workflowNodes 3 workflowXLabs Sampler workflowAdetailer workflowoutpaintflor workflowfaceswap2 workflowfaceswap3 workflowREACTORFACESWAP workflowpullflo workflowpulidtxt workflowpluidlux workflowfillparam workflowPULIDREDUX workflowREDUXOTHERPARAM workflowshaker workflowLiveparam workflowLIVEPORTRAITIMGPARAM workflowLIVEPORTRAITVID2VIDPARAM workflowLTXPARAMTEXTOVIDGGUF workflowLTXPARAMTXTFP workflowLTXIMAGETOV workflowhuynyav2vparm do you have any hint for how to resolve either? Happy with either fixing the combined comfy flux + hidream or just having the flux only restored (since I find the flux results superior to hidream)
Paul Mazurkiewicz
Hey AIT, I tried substituting in the full model instead of the DEV model (Quant 8) and it failed...just produces garbage. What else do I have to change in the text to image GGUP or FP8 workflows to use the full model instead of the DEV model? Is there a difference in quality?
Bobby Boe
Hi my friend. I tried to use your workflow, but one node "workflowloader" refuses to get installed. I wonder why. I have Hidream in another workflow running, but was curious about your workflow. Is that "workflowloader" essential? could it be substituted? I am running the desktop-version of ComfyUI, in which it is a little complicated (for me) to install via pip... . so usually I try to do everything using the manager. (yes, I updated everything). What package does the missing node "workflowloader" belong to? I am using 4060TI 16GB, so I would use the right GGUF-side. ------------------------ Actually now, I saw, this node is the image-loader in your img2img-workflow. In the worflow the name was "workflowloader" (and red because missing), and in your YT-Video, it is only called "loader". I can see that is has some more options than a normal image-loader. Can you tell me what package it is from, and if the additional options are really needed (and why?)
Aitrepreneur
open the manager and update all the nodes
Niklāvs Trankalis
Where can I find the workflowhiinpai nodetype? It's the only Missing Node Type.. everything else seems to be fine
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)?
Aitrepreneur
What are you talking about?
Li
the attached json is not at all as described. It is very simple. Please give us the right file
