Step 1: Upload the photo of the character you selected.
Step 2: Describe the video you want to generate, for example what actions the person performs, the camera movement path, and any background changes.
Forget the hassle of installing nodes—just try it online first!
🎁 Use my invite code rh-v1213 to grab 1,000 free points (enough for ~20 videos).
👉 https://www.runninghub.ai/post/1957365514316214274
Description
Added workflow for local decoding of latent files
FAQ
Comments (11)
Hmpf, I am very sorry. I found the latent file under "output". Now it's another workflow version without this latent file. Don't you need it anymore?
On ur local machine u dnt need the latent file, Only when generating NSFW videos online, and to evade censorship, are latent files downloaded locally for decoding
Is there a portable version with comfyUI?
If you want to use ComfyUI on your phone, check this out: https://www.runninghub.ai/ai-detail/1958201161302085633?inviteCode=rh-v1213 This is the workflow I made on PC and the generated app interface, which can be easily used on a phone.
For starters thanks for this amazing workflow. It works like a charm.
But I am still confused about where the "comfyui/input" folder is? I searched everywhere in workflow and even on comfyui. Sorry I am new to this so not aware.
This folder is on your personal computer, located inside the "ComfyUI" folder of your Comfy startup directory.
WanVideoSetLoRAs
Set LoRA node does not use low_mem_load and can't merge LoRAs, disable 'merge_loras' in the LoRA select node.
Hello, I got this error (Im using RTX 4070 Ti). Do I need to disable merge lora? I tried disabling it and the process is very slow (2 hrs and nothing happening. I'm using every file exactly same as the workflow btw.
I have the same issue and just set merge LoRA to false, it runs, produces videos but colors are wrong, I wonder is this the reason or I have something else wrong? I have 3090Ti
I found what is the issue with colors - start_steap and end_step are not set properly for WanVideoSampler, duh
Not working because sarge attention + triton cannot be installed on latest comfyui that needs python 3.13. and when i use SDPA then i got wrong color results.
It also seems that this WF is just promotion for your referal advertising here... ;)
I had issue with colors though I managed to install sageattention on my comfy, looks like it is not related to attention - start_step and end_step are not set properly for WanVideoSamplers