Searge-SDXL: EVOLVED v4.3.2
Version 4.x is here.
I made a convenient install script that can install the extension and workflow, the python dependencies, and it also offer the option to download the required models.
Please read the install instructions in the readme file for more information about the install script and how to use it.
Description
Some custom nodes for ComfyUI and an easy to use SDXL 1.0 workflow. A detailed description can be found on the project repository site, here: Github Link
Installing
Detailed install instruction can be found here: Link to the readme file on Github
Usage Notes
The workflow is in the workflow folder, please load it from the JSON files to make sure you are using the latest versions of it.
The images in the examples folder have workflows embedded.
How to use this workflow
Here is an amazing Youtube video that Nerdy Rodent made about version 3.4 of the workflow. Many things have changed in v4.0 but the basics still apply. And while you are watching it, don't forget to subscribe to the channel.
Link to the Nerdy Rodent video
Updates
What's new in v4.3.2?
This is a minor update to make the workflow and custom node extension compatible with the latest changes in ComfyUI.
What's new in v4.3.1?
This is a minor update to make the workflow and custom node extension compatible with the latest changes in ComfyUI.
What's new in v4.3?
This update added support for FreeU v2 in addition to FreeU v1.
New Features
Support for FreeU v2 has been added and is included in the v4.3 workflow
Added more presets for FreeU and a selector to switch between v1 and v2
Updated the example images to embed the v4.3 workflow
What's new in v4.2?
This update contains bug fixes that address issues found after v4.0 was released.
Bug Fixes
A recent change in ComfyUI conflicted with my implementation of inpainting, this is now fixed and inpainting should
work again
New Features
Support for FreeU has been added and is included in the v4.2 workflow
Note: the images in the example folder are still embedding v4.1 of the workflow, to use FreeU load the new
workflow from the.jsonfile in theworkflowfolder
What's new in v4.1?
This update contains bug fixes that address issues found after v4.0 was released.
Bug Fixes
The high resolution latent detailer was not properly set up in the processing pipeline and did nothing
The debug printer node was broken - I didn't notice that because it was not connected in any of the v4.0 workflows
A bug related to generating with batch sizes larger than 1 has been fixed, it's now working properly
Other Changes
The images in the
examplesfolder have been updated to embed the v4.1 workflow
What's new in v4.0?
A complete re-write of the custom node extension and the SDXL workflow
Highly optimized processing pipeline, now up to 20% faster than in older workflow versions
Support for Controlnet and Revision, up to 5 can be applied together
Multi-LoRA support with up to 5 LoRA's at once
Better Image Quality in many cases, some improvements to the SDXL sampler were made that can produce images
with higher qualityImproved High Resolution modes that replace the old "Hi-Res Fix" and should generate better images
... many more things, too many to mention them all here
Description
What's new in v3.2?
More prompting modes, including the "3-prompt" style that's common in other workflows (using separate prompts for the 2 CLIP models in SDXL (CLIP G & CLIP L)and a negative prompt)
3-Prompt G+L-N - Similar to simple mode, but cares about a main, a secondary, and a negative prompt
and ignores the additional style prompting fields, this is great to get similar results as on other
workflows and makes it easier to compare the imagesSubject - Style - The subject focused positives with the style focused negatives
Style - Subject - The style focused positives with the subject focused negatives
Style Only - Only the positive and negative style prompts are used and main/secondary/negative are ignored
Weighted - Overlay - The positive prompts are weighted and the negative prompts are overlaid
Overlay - Weighted - The positive prompts are overlaid and the negative prompts are weighted
Better bug fix for the "exploding" the search box issue, should finally be fixed (for real) now
Some additional node types to make it easier to still use my nodes in other custom workflows
The custom node extension should now also work on Python 3.9 again, it required 3.10 before
FAQ
Comments (13)
The first workflow that are very useable.
Thanks for all your work !
Is adding face detailer in your workflow in your to-do-list ?
It's somewhere on my ideas list, but I'll try something else first to see if that also works to improve faces.
Excellent workflow, I'm now using it instead of the personal one I'd cobbled together. The layout is much clearer in this version and it's easy to use. :-)
I like that it plays nicely with the Impact Pack's WildcardProcessor.
I echo seto007's suggestion. I've turned the base ratio right up and the refiner strength right down as I'm using Dreamshaper (and connected the Base Model VAE output instead of the VAE Model output).
One minor niggle is that a lora strength range of 0.0->1.0 is hardcoded, but some SDXL loras suggest strengths of -5 to +5. It's easy enough to edit the .py file to change that but some people might not be comfortable with that.
Thanks for the feedback. And the suggested value range for the lora strength makes sense. I have to admit that I didn't even think about it too much, because I rarely use any loras and right now.
I'm still focused on understanding better what what SDXL base & refiner can do on their own, without switching to a custom trained model or applying any loras.
I'll adjust that min and max value in a future update, so that nobody is forced to make changes to the python scripts.
You need to update ComfyUI to the latest version
Does anyone know where I could read about the differences or relationship between "base resolution", "Conditional Scale Factors", "Upscaler Resolution Scale Factor" and "Hires Fix Scale Factor"?
EDIT: No worries! I could make much more sense out of it now that I found the message in the WF to connect nodes. I didn't notice this at first,
"^ Connect to enable upscaling
| ---------------------------------
| Connect to enable saving of image
v after hires fix is applied"
any videos on your work flow, please. I am just a beginner, so I don't understand how to use it fully.
Unfortunately I'm not a Youtuber, so I don't make my own videos. Until one of the Youtubers who covers AI image generation picks it up for a video, you'll have to resort to my written documentation in the readme file.
I'm extending the documentation more and more going forward.
@searge slowly and steadily, I am getting hang of it but my only problem is that.
how to use the specific seed If I create five pictures and want to use the fifth picture seed, how do I do that?
Also, I have noticed if I keep the same prompt, ComfUI keeps producing the same pictures. Randomizing the seed doesn't change it, so what do I do?
If you generate multiple pictures with different seeds, they are all saved in a folder in the Comfy output directory. You can load the png of an image you like in ComfyUI and it will get back the workflow, settings, and seed that was used to generate that image.
After you load that image, I recommend setting the seed to fixed so that it won't change the very moment you generate the image from it again.
Nerdy Rodent made a video about the workflow: https://www.youtube.com/watch?v=_Qi0Dgrz1TM
@searge already gotten a hang of it don't need any video, but thank you man. now only if control net was available.