NEW ANIMA VERSION! (don't need the 'nsfwupscaler' tag anymore)
SETTINGS: Trained on anima base, results will vary on other models. Use anima base model suggested settings.
Excited to bring you the Anima version of this lora!
I used the same dataset/tags as the popular pony version!
All sample images were made on BASE without any upscaling or other enhancements!
I REMOVED THE 'nsfwupscaler' tag completely and kepth the typical wd14/danbooru tagging used for the pony version.
I wrote custom code to fix a cfg issue I found in the training loop of kohya ss and trained locally for 2040 steps with great results!
I feel like this lora is worthy or following up the old one. I'm looking forward to seeing what people come up with.
**The anima version does not have a workflow. I used a custom sd script I modified in kohya ss to quickly generate the sample images. Settings are accurate, used the er_sde sampler and beta57 scheduler (custom added, pulled from comfyui code).
PONY VERSION:
trained on nsfw photos to increase quality for realistic images. the string 'nsfwupscaler' was included in all captions, however this is intended for normal use, not just upscaling.
Description
FAQ
Comments (16)
thank you but where is the workflow?
download the training data file (you'll see it's named 'pony-realism-22-nsfw-photo-comfyui-workflow.zip')
This seemed to do nothing for me - even made things less realistic. But I am using cyberrealisticPony_v65 - so maybe it's already realistic enough.
This was trained for the specific version of pony realism, using that model during the training, so it's likely that it will have mixed or poor results with other models
Delicious
Can I add a facial detailer to this workflow? What would be the best setup?
I'd add an entire separate section with clipseg or similar to identify/mask the face then use a model to do i2i on the masked out face before passing to the final/upscale stage
Thanks! I can't seem to get Comfyui to recognize the Lora file.
LoRA Stacker: - Value not in list: lora_name_1: 'SDXL_lora/pony-realism-22-photo-lora_mk3/pony-realism-22-nsfw-photo.safetensors' not in ['None', 'SDXL_lora\\pony-realism-22-nsfw-photo.safetensors', 'pony-realism-22-nsfw-photo.safetensors']
Looks like it needs to be re-selected from the list as your download isn't in the same folder structure
new to this. im using comfyui. where do i place the training data
The training data is a zip containing the workflow. Download, extract, drag and drop into comfy
@CautionMergeAhead thank you for the help sorry it took so long to respond
i'm getting an error about missing node 9?
you can open the json file to see which node is node 9.
search for: "id": 9
find: UltimateSDUpscale
Search UltimateSDUpscale in Custom Nodes Manager
If you don't have Comfyui Manager, you should get it.
Follow instructions to install here: https://github.com/Comfy-Org/ComfyUI-Manager
I suggest you use my new model instead of this old lora though:
https://civitai.com/models/2151914/verisimilis-photoreal
And consider my new loras, using standard current workflows for illustrious with high-res fix and upscale stages (and even inpainting if you want to get fancy):
https://civitai.com/models/2037809/realism-detailer-refiner
https://civitai.com/models/2001172/realistic-pussy-detailer-experimental
he descargado el flujo de trabajo y lo he puesto en el lienzo he descargado los nodos desde el manager y sigue sin funcionar , soy nuevo en comfyUI y me esta costando un poco jajaja..
This is pretty old now. Just Use a default two stage illustrious workflow using pony realism and this lora and add nsfwupscaler, to the prompt.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.