About
Produce real life versions of
anime
cartoon
3D images
or enhance quality of photorealistic images
Downloadable zip contains the general purpose json workflow, and example images with optimized settings for different kinds of images. Open the image in ComfyUI to get the optimized setting workflow.
Or download an example image above and open it in ComfyUI.
Instructions
1. Load Image
2. Press Run
3. Done! (~10s on a 5090)
Note: The default settings and lora strengths are fine for most images, but experimentation with lora / weights to find sweet spot for your particular image. Best use case is for flat 2D cartoons. anime; also effective for enhancing existing photorealistic images. Partial 2.5D images seem to need the most setting tweaking.
Options
Resize image
Default: true; resize to 1024 max length
[Optional] Set max side length; maintains proportions
Set false; Uses original image size
Join before/after images into one
Default: true; joins before | after - L | R
Set false; no output
Diffusion model
LoRAs
(optional; great for nsfw quality, may add human faces to animals, may oversmooth)
anime2real-semi (may add human faces to animals, may oversmooth)
Special thanks to
Description
About
Produce realistic adaptations of
anime
cartoon
3D images
Enhance quality of photorealistic images
Instructions
1. Load Image
2. Press Run
3. Done! (~10s on a 5090)
Note: The default settings and lora strengths are fine for most images, but experimentation with lora / weights to find sweet spot for your particular image. Best use case is for flat 2D cartoons. anime; also effective for enhancing existing photorealistic images. Partial 2.5D images seem to need the most setting tweaking.
Options
Resize image
Default: true; resize to 1024 max length
[Optional] Set max side length; maintains proportions
Set false; Uses original image size
Join before/after images into one
Default: true; joins before + after = L|R
Set false; no output
Diffusion model
LoRAs
(optional; great for nsfw quality, may add human faces to animals, may oversmooth)
anime2real-semi (may add human faces to animals, may oversmooth)
FAQ
Comments (14)
can't install flownodes, the node rezize image/logic if is missing
to custom_nodes
"git clone https://github.com/silveroxides/ComfyUI_LogicMath.git"
git checkout 0d77b4e4f4c1120afaede9ef9a3668aa72b582ae
@n12087377675 Who are you sir???
Nothing happens after I click run.
Prompt executed in 0.14 seconds
got prompt
Failed to validate prompt for output 197:
* Int Compare 180:
- Value not in list: comparison: '0' not in ['<', '>', '==']
- Failed to convert an input value to a INT value: a, >, invalid literal for int() with base 10: '>'
Output will be ignored
Failed to validate prompt for output 172:
Output will be ignored
Failed to validate prompt for output 175:
Output will be ignored
Prompt executed in 0.18 s
It looks like you didn't add an input image
@ultimo_intento
I did load an image.
Failed to validate prompt for output 197:
* Int Compare 180:
- Value not in list: comparison: '0' not in ['<', '>', '==']
- Failed to convert an input value to a INT value: a, >, invalid literal for int() with base 10: '>'
This would imply that one of the input values was 0 or otherwise not interpretable. My guess is that either there was an issue loading the image (e.g. corrupted) or an incompatible version of the node. Try with other images or try reinstalling the node, then ask ChatGPT if still cannot resolve the error in your comfy environment.
@ultimo_intento
chatgpt was useless. tried other images, same. switched from latest version of comfyui-workflow-encrypt node to v1 and to nightly. Same.
@drfaker911219
ಠ_ಠ workflow encrypt node?
smh
cnr_id : flownodes
ver : 8858ca94d2c4af194e913d7a63322861fc1ca3a3
@ultimo_intento
Thanks it works now
Logic is missing...
i dont get the same results as yours, do you have a video tutorial?
Instructions
1. Load Image
2. Press Run
3. Done! (~10s on a 5090)
Note: The default settings and lora strengths are fine for most images, but experimentation with lora / weights to find sweet spot for your particular image. Best use case is for flat 2D cartoons. anime; also effective for enhancing existing photorealistic images. Partial 2.5D images seem to need the most setting tweaking.
It's literally that simple :)