finished up img2img 1 todo left, applied all changes to txt2img 1 todo left

workflows/IMG2IMG/v4.4/changelog.md

@@ -1,6 +1,6 @@
 - implementation of new image loading node (by me) to be able to multi-select one or more images for a run (will be executed one after another)
 - dynamic prompts/wildcards now supported through Impact-Pack, syntax [here](https://github.com/ltdrdata/ComfyUI-extension-tutorials/blob/Main/ComfyUI-Impact-Pack/tutorial/ImpactWildcard.md) - txt files have to be in the folder BEFORE comfy starts to be recognized
-- introduction of refiner capabilities
+- introduction of refiner functionality
 - switch for turning off seperate VAE to use baked-in VAE of checkpoint
 - replaced the single lora loaders and their 6 activation buttons with rgthree's lora stack loader to load as many as you want (same for detailer loras)
 - moved all model selection stuff into second row for more streamlined process
@@ -8,6 +8,6 @@
 - fixed v-pred model bug that didn't apply v-parameterization
 - fixed modelname not saving correctly for civitai
 - overhauled notes in the workflow for better understanding & improving your results
+- global clipskip now happens AFTER lora loading (correct implementation, no difference to outcome quality)
 
-_low VRAM for transfer not put into a subggraph yet because of an output bug when bypassed in current comfy versions_
-_wanted to add recommended detailer values you can just activate/deactivate (more vram but better results) but this is currently not possible since multiple custom_nodes still cause issues with subgraphs_
+_wanted to add recommended detailer values you can just activate/deactivate (more vram but better results) but this is currently not possible since multiple custom_nodes still cause issues with subgraphs being bypassed_

workflows/IMG2IMG/v4.4/todo.md

@@ -0,0 +1 @@
+- add v-pred & clip skip to refiner

workflows/TXT2IMG/v4.4/changelog.md

@@ -0,0 +1,11 @@
+- dynamic prompts/wildcards now supported through Impact-Pack, syntax [here](https://github.com/ltdrdata/ComfyUI-extension-tutorials/blob/Main/ComfyUI-Impact-Pack/tutorial/ImpactWildcard.md) - txt files have to be in the folder BEFORE comfy starts to be recognized
+- introduction of refiner functionality
+- switch for turning off seperate VAE to use baked-in VAE of checkpoint
+- replaced the single lora loaders and their 6 activation buttons with rgthree's lora stack loader to load as many as you want (same for detailer loras)
+- moved all model selection stuff into second row for more streamlined process
+- moved controlnet & ipadapter model selection into their respective groups for easier switching
+- removal of all group nodes & abstraction through subgraphs
+- overhauled notes in the workflow for better understanding & improving your results
+- global clipskip now happens AFTER lora loading (correct implementation, no difference to outcome quality)
+
+_wanted to add recommended detailer values you can just activate/deactivate (more vram but better results) but this is currently not possible since multiple custom_nodes still cause issues with subgraphs being bypassed_

workflows/TXT2IMG/v4.4/todo.md

@@ -1,10 +1 @@
-TODO:
-- add dynamic prompts compatibility, S&R {red|blue|green} etc.
-- extra boolean for baked in vae that inject vae from model loader instead
-- implement a refiner
-- check if v-pred model works correctly
-- swap group nodes for subgraphs, turn big major tasks into subgraphs
-- swap lora loaders
-- check if modelname saves correctly
-- overhaul all notes
-- overhaul full design, same philosophy as IMG2IMG workflow (ControlNet to ControlNet Group, IPAdapter same)
+- update how to note & debug note