ComfyUI: LINK
Github repo + nodes description: LINK
Leave suggestions and errors if you meet them
What's new in 0.5.0:
CombiningArea scaler
More user-friendly ui names
ALL nodes description moved to GitHUB
Tuples and so on moved to their own directory in UI
Simple* introduction:
Automate calculation depending on image sizes or something you want
easier(or not) editing multiple values of various nodes
Math
Modded scalers
Installing: unzip files in ComfyUI/custom_nodes folder
Should look like this:
For example (v0.5.0) there is an example how scaled ConditioningArea can improve image after scaled latent combining:
Only LatentCombine:
Combining preview:
LatentCombine with scaled ConditioningArea (640*360 to 1360*768):
Example of workflow i made for this located in: /Derfuu_ComfyUI_ModdedNodes/workflow_examples/
model: hPANTYHOSENEKO (sorry, couldn't find link)
negative promp: embedding:verybadimagenegative6400
TROUBLESHOOTING:
If there are troubles with different sizes, aside from *64, this may solve problem: found on GitHUB
This code is at the end of this file: /ComfyUI/comfy/ldm/modules/diffusionmodules/openaimodules.py
NOTES#2:
Debug nodes counts as OUTPUT nodes and can be used withowt image preview or save nodes to get results
P.S.:
All fixes wou can find or post on github, i look there too
If you catch error like: Calculated padded input size per channel: (2 x 82). Kernel size: (3 x 3). Kernel size can't be greater than actual input size. This MAY be because of too high or low offset you give to node
Description
Included workflow example
Created git repository
Structure changes
Added:
ABSNode
LatentCompose (MODDED)
RandomFloat
FlipTuple
DebugNodes