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. If you have any ideas for LoRA's let me know. I only have an RTX 2060 6gb so they might not be the best, the max resolution that I can train is around 600x600.
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
This version has minor improvements.
I made lora for skin textures and merged into Eden. Should give better skin, you can use "detailed skin, brown skin, dark skin, veins, veiny skin, pores" to enhance the effect.
I also made a lora for photography and merged it. this should improve the realism of landscapes, you can use
landscape
portrait - for upper body generations
full shot - should give you the full subject
Please let me know if this breaks anything as well as suggestions.
If anyone wants it I may try and train a lora for other skin colors (like blue and green) and see if works.