Nodes/ComfyUI-Apt_Preset/Image_effect_Load
ComfyUI Node

Image_effect_Load

Preview four filter presets on one image at once

By cardenluo·Created 2 years ago·Updated 18 days ago· 309
Image_effect_Load
  • image_input
  • output1
  • output2
  • output3
  • output4
image
output_01_fxoriginal
output_02_fxgrayscale
output_03_fxflip_h
output_04_fxflip_v

A quick way to see several stock filters applied to the same source image side by side, instead of wiring a separate node per effect and running the graph four times. Load - or feed in - an image, pick up to four effects from a dropdown, get four filtered outputs back from one node.

How it works

image is a file picker like ComfyUI's core LoadImage node - it pulls from your input directory - with an optional image_input override if you'd rather feed it a wired IMAGE from earlier in your graph instead of loading from disk first. output_01_fx through output_04_fx each choose one effect from a roughly 30-item list; the defaults are original, grayscale, flip_h, and flip_v. What the brief shows beyond that - enhance, posterize, sharpen, contrast, equalize, sepia, blur, emboss, plus 18 more - is enough to confirm it's a real preset library, not a handful of toy filters, but the full list isn't documented anywhere outside the live dropdown, so check ComfyUI itself for the complete set rather than trusting a partial one here.

The inputs and outputs that matter

  • image (a file dropdown) or image_input (optional IMAGE override - use this if your source is already in the graph).
  • output_01_fxoutput_04_fx - four independent effect choices.
  • Output: output1output4, one IMAGE per effect slot, in order.

How to install it

Via ComfyUI Manager: search ComfyUI-Apt_Preset, install, restart. Manually:

cd ComfyUI/custom_nodes
git clone https://github.com/cardenluo/ComfyUI-Apt_Preset.git

Windows: install.bat. Linux/Mac: pip install -r ComfyUI-Apt_Preset/requirements.txt in your venv. Restart ComfyUI. Nothing to download for this node - the effects are computed locally, no external model - though the pack's requirements.txt (pandas, gguf, onnxruntime, transparent-background, scenedetect, and more) is heavier than a filter node alone would need, because Apt_Preset is one large, actively maintained pack (~300 GitHub stars) covering a lot of unrelated ground.

Common issues & troubleshooting

No fine control over any single effect. This is a preset picker, not a parametrized filter node - if "blur" is too strong or "sharpen" isn't strong enough, there's no radius or amount slider here. Use this node to figure out which stylistic direction you want, then reach for a dedicated node (blur, sharpen, etc.) once you know.

image_input seemingly ignored. Make sure you're not relying on the image dropdown selection at the same time you're wiring image_input - the override exists specifically so you don't have to save a file to disk first, but double-check which one is actually driving the node if the output doesn't match what you expected.

Node missing after an update. Every Apt_Preset node ships from one Python module - a broken dependency anywhere in requirements.txt can hide the whole pack from the menu, not just this node. Check the ComfyUI console for import errors first.

CategoryApt_Preset/image/color_adjust

Inputs (6)

NameTypeDefaultDescription
imageCOMBO1 options: example.png
output_01_fxCOMBOoriginal30 options: original, grayscale, enhance, flip_h, flip_v, posterize, +24
output_02_fxCOMBOgrayscale30 options: original, grayscale, enhance, flip_h, flip_v, posterize, +24
output_03_fxCOMBOflip_h30 options: original, grayscale, enhance, flip_h, flip_v, posterize, +24
output_04_fxCOMBOflip_v30 options: original, grayscale, enhance, flip_h, flip_v, posterize, +24
image_inputoptIMAGE

Outputs (4)

NameTypeDescription
output1IMAGE
output2IMAGE
output3IMAGE
output4IMAGE