This is a very simple ComfyUI beginner friendly Image-to-Image workflow that will work with a single WAN 2.1 GGUF model with multiple LORAs (This workflow does require ComfyUI's LORA Manager plugin to function. It is a good idea to install both ComfyUI manager and LORA Manager plugin to help you easily download and manage Checkpoints, LORAs and other resources. Not only these two are helpful for this workflow but they will help you a lot in any other cases). You need a Hugging Face account to download your necessary WAN 2.1 files (Details are mentioned below). Make sure you install GGUF addon for ComfyUI using ComfyUI manager and place the correct files in correct places. Also check out my other workflows for SD 1.5 + SDXL 1.0, SD 3.5, HiDream and Flux.
How to use this -
#1. Just select your desired WAN 2.1 GGUF model first and then load
#2. one or multiple matching LORA(s) from LORA Manager to start
#3. then select an input image to write your positive and negative prompts for output.
#4. select how many images you want (Change the number besides the "Run" button)
#5. select image sampling methods, CFG, steps etc. settings
#6. finally press the run button to generate. That's it..
*** This workflow will also function if you know how to properly disable the LORA Node and bypass it, and I think with minor tweaks it can be transformed into a viable Image-to-video workflow for GGUF model with LORA support.
Enjoy!
### To use this workflow you need to log into huggingface and download necessary files from there (I also included a text file on the archive that has the workflow file, in which you will find even more links for essential downloads for my other workflows) -
# Download Links for WAN 2.1 Checkpoints
# Download Links for WAN 2.1 Encoders
https://huggingface.co/city96/umt5-xxl-encoder-gguf/resolve/main/umt5-xxl-encoder-Q3_K_S.gguf
# Download Links for WAN 2.1 VAE
### I also included the input image and the output (as seen in the workflow preview image on this workflow's CIVITAI Page) on a second zip file for reference
Well CIVITAI won't let me upload second zip file but I uploaded both images in this workflow's gallery.. if you want to use the input image feel free to download it from that gallery. Also same image can be found in my another resource's gallery.
=========================================================================================================================
These are standard / native aspect ratios supported by this workflow that are optimized for best performance and quality:
Dimensions (W×H) Aspect Ratio
Square 1024x1024, 1440x1440 1:1
Portrait (classic) 1024x1536, 1280x1920 2:3
Portrait (Web) 896x1120 4:5
Portrait (tall) 1080x1920, 896x1600, 720x1280 9:16
Wide Photo (old) 1536x1024 3:2
Widescreen 1920x1080, 1600x900, 1280x720 16:9
Description
FAQ
Comments (6)
WAN 2.1's Image generation is very good, with very few steps it can do images with perfect texts. Just wished it had it's own easy methods to do Depth, Pose, Canny etc. like Flux. I love how closely it follows the prompt.
errr... it DOES! Ever hear of VACE?
blobby99 yeah I am looking into VACE and also WAN 2.2 5B Q models, these are very interesting.. I also came up with WAN GGUF image inpainting and outpainting workflows by tweaking ComfyUI's basic SD inpainting and outpainting models you can check them amongst my CivitAI model uploads.. I am liking WAN bit more than Flux currently as it is faster and easier for images, also it can do good videos almost without any deformed limbs or weird movements compared to others.. wished CivitAI should offer ways to create WAN loras.
The lora manager doesn't see my loras
You have to configure the path if you installed LORA in a custom path.. I simply keep things simple and tidy with my setup.. unlike other plugins this does allow me to download and organize LORAs by the LORA Manager itself.. a workaround that will work for you would be to download any other LORAs you may not have now using LORA Manager.. find and copy the new path (or set a new path completely chosen by you) and relocate older LORAs that you have collected to the new path.. I usually keep my LORAs and Checkpoints stored inside subfolders that exactly lets me know what type of base model I am using.
@sarcastictofu I apologize for being helpless. I read the repo, configured the api and did my due diligence. Thanks for the workflow, its running now.
My post was a cry for help really, a simple whisper into the ether, "is there someone out there?"
Anyway, do you have any recommended loras that specifically increase the influence of the source image onto the output?
Ty