High-Res-Fix and Dynamic Thresholding
This model will no longer be updated as I feel like its starting to regress, I will be training LoRA's from now on.
This model's goal is to generate photo like images.
Inpainting version on HuggingFace
All images where generated using High-Res-Fix with 4x-UltraSharp upscaler.
vae-ft-mse-840000-ema-pruned
Description
Please let me know if this breaks anything as well as suggestions
If the effect is too much, try playing with the weight of skin lora trigger words.
Fantasy skin colors (red, blue,green), seem to be hit or miss, from my testing you will see more success if you add words for other things that also have that skin color (ogre, avatar)
After more testing I found that my photography lora was changing generations too much. I have removed it. In this version I retrained my skin lora and mixed some NSFW and anime/fantasy models.