Hi ender570,
For those who wish to use their own ComfyUI environment, please refer to this reply as an initial reference. A more comprehensive document will be available soon.
Hi LairdPaul
Sharing this one with you for reference, too.
After a few years out of iclone i just bought iclone 8 and character 4 (and waiting for version 5). I have benn using AI for a few years already and i think is a valuable tool. This AI renders seems a great idea, i have benn mixing AI with 3d for a while. I want to use this, but having ComfyUI running remotely on Runpod. I already do use it on Runpod. What i wanted to know is how to do it with this plugin? Is there a tutorial, anybody can help me, I really want to try it but i dont want ComfyUI running locally, its too slow. 
thanks
Hello, can someone tell me where I can find the RL_ComfyUI folder?
Greetings, Robert.
It’s installed in C:\Program Files\Common Files\Reallusion\RL_ComfyUI
But if you want to run it, navigate to http://127.0.0.1:8188/ in browser when server started
Many Thanks 
Greetings Robert
iClone render result:
iClone AI render result:
In AI render result, the actor motion did not follow the actor original motion and looks like the iclone camera moving action also did not followed in AI render result.
When using AI rendering video, the “Wan2.1 Fun 1.3B Control” always render with error. The “Wan2.1 Vace 1.3B” can render video but with error messge “Executio failed”, but you can see in the ComfyUI, it is still rendering the video.
Below is imported Reallusion provided “img2img_Realistic_Photorealism_Image_Style.json” workflow into AI image rendering. It is working and the result is OK.
One more thing, the controlnet “Pose” sometime can work, but most of time cannot work. Do not know why. Maybe it is related to the image or characters.
Do you know how can i work with it using Runpod? Thanks
I have no idea… sorry.
I posted this in the bug thread, but I wanted to see if anyone else had this problem.
I just installed the AI Render on my laptop. As soon as I move my mouse over the UI it disappears. So I can’t connect to ComfyUI. Any idea why this may be?
My system is as follows:
| Device name | BlackHawk |
|---|---|
| Processor | 13th Gen Intel(R) Core™ i9-13950HX (2.20 GHz) |
| Installed RAM | 32.0 GB (31.7 GB usable) |
| Device ID | D89F1245-140D-4333-B396-BBF79608BB52 |
| Product ID | 00355-60883-91766-AAOEM |
| System type | 64-bit operating system, x64-based processor |
| Pen and touch | Pen support |
Hi mecolef,
The reason OpenPose may not be functioning is that your character is of the Humanoid (non-standard) type. Support for Humanoid is currently under development.
Pose tracking currently supports CC Standard and ActorCore characters (only CC3+ characters include facial data).
Hello, mecolef,
For the Wan2.1 Fun 1.3B Control case, please go to your history folder, run the Workflow_Gen.json file in ComfyUI, and check whether it works. If it doesn’t, please share a screenshot of ComfyUI showing the issue.
For the Wan2.1 Vace 1.3B case, it appears the process failed on the iClone / Character Creator side but continued running in ComfyUI. We will investigate this case as soon as possible.
Sorry for the inconvenience.
Hello, blackhawk,
Please allow me to confirm your case.
Is it correct that you can open the Settings Panel, but all fields except the IP address disappear when you move your mouse over it?
Hi Maelyn_RL,
Another bug, when rendering a video, it always shows a file saving dialog for saving the video, but it cannot save the video even you type in correct path.
When using “wan2.1 Fun 1.3B Control”, the ComfyUI terminal will below error message:
That is correct. Everything will disappear on the UI. I can move my cursor over the top of the UI and everything will disappear except the green strip at the top allowing to close the plugin. Otherwise I cannot connect to ComfyUI.
Testing Wan2.2 Fun Model, trained on iClone “SuperRob”.
SuperRob in his cosy Influencer-room. 
iClone Camera Tests in Comfy :
5 Likes
I still have not gotten to the bottom of LoRA. Nonetheless, this one came out fairly well.
(background was added through depth mask in NLE, as well as camera shake 
)
4 Likes
Hello, mecolef,
Thanks for reporting this issue.
I saw the error message you provided, and it appears to be related to the Python Torch library. Here are a few points I’d like to confirm:
-
ComfyUI version – Are you using RL_ComfyUI or your own ComfyUI setup?
If you are using RL_ComfyUI, please make sure the Torch and TorchVision versions are:- torch-2.7.1+cu128
- torchvision-0.22.1+cu128
These should be located in: RL_ComfyUI\python_embeded\Lib\site-packages
-
Image resolution – What resolution are you using? Some workflows require specific sizes.
We have tested 512×512 and 1280×720, and both worked well. -
Custom AI Render Tool – Did you create your own AI Render Tool that uses your custom workflow as image workflow and our template as video workflow?
If so, the output image might be in an invalid format for our template’s video workflow. -
About save video – Do you mean the video is generated successfully in the history folder but not saved to the location you selected through the file dialog? Based on the issues above, it seems that the video generation workflow is currently not working properly on your case. If the video is not generated successfully through our workflow, it will not be found in either the history folder or the specified location.
If it’s convenient for you, please send me your history folder and project via a private message on the forum (click my profile photo, then use the Message button at the top-right).
Many thanks.
Hello Melvin_RL,
Noted, hope the new tracking for Humanoid, or non-standard characters will soon supports!
Thanks for your advice!