For high-quality AI Image prompts, check out my website👉 https://promptdexter.com
LoRA designed to reduce the typical smooth/plastic AI look and add more natural skin texture and realism to images. It works especially well for close-ups and medium shots where skin detail is important.
You can also use for upscaling too or restoring old photos, which actually intended for.
Version Comparison
OVERVIEW = V4: More quality + NSFW detailing support, V3: Professional/Celebrity Skin, V2: Realistic Natural Skin, V1: Too Much Details
V4 is trained on more refined and 50% nsfw dataset, so besides better quality it has NSFW detailing capabilities too. Note: This is only for detailing, for simple nsfw use nsfw words in prompt, for hardcore stuff use other loras with it.
V3 is trained on professional 4K quality images with higher steps. It applies very subtle skin texture which looks more realistic, without producing freckles and burnes.
V2 for more natural looking skin texture than V1, can produce freckles and moles for some time. It is good at preserving skin tone and lighting also.
V1 tends to produce overdone skin texture like more pores and freckles, and it can change lighting and skin tone also.
🛠️ Editing Workflow - https://github.com/vizsumit/comfyui-workflows
V4 Weight: 0.55, Prompt: Make this photo high-quality. Preserve subject's face, expression, figure and pose. Preserve overall composition of this photo.
V3 Weight: 0.6, Prompt: Make this photo high-quality. Preserve subject's face, expression, figure and pose. Preserve overall composition of this photo.
V2 Weight: 0.6, Prompt: Make this photo high-quality with natural skin texture. Preserve subject's facial features, expression, figure and pose. Preserve overall composition of this photo.
V1 Weight: 0.5, Prompt: Make this photo high-quality featuring realstic skin texture. Preserve subject's facial features, expression, figure and pose. Preserve overall composition of this photo.
🖼️ Generation Workflow - https://github.com/vizsumit/comfyui-workflows
Weight: 0.7-0.8
V4 Prompt: This is a high-quality photograph of a (your subject...)
V3 Prompt: A high-quality photograph of a (your subject...)
V2 Prompt: A high-quality medium close-up photo with realistic skin texture featuring a (your subject...)
V1 Prompt: This is a high-quality photo with realistic skin texture and details of a (your subject...)
Feel free to try it and share results or feedback. 🙂
Description
FAQ
Comments (13)
Awesome LoRA! It really does fix the overly-polished look that Flux.2 Klein tends to produce.
Glad you like it. 🍻
I don't know what to say. There are too many freckles. 0.5 or 0.3 nothing changes.
Most of the time it doesn't add freckles, but for some skin types it does. To avoid it add "without freckles, without moles" in your prompt.
@vizsumit pos or neg prompt?
@carlitajohn89931 the way he worded it sounds like you use it in the positive prompt. For a negative prompt you would say "with freckles, with moles" because negatives take the inverse.
Flux.2 does not use negative prompts, and with cfg at 1.0 it ignores it anyway (cfg 2.0 and higher uses a negative prompt). You only create longer rendering times, so stick to cfg 1.0
Mate!!! This is the best lora for skin. I am so blown away by how good this lora is. I can take a plastic Qwen picture and she turns from a poor sex doll into a very realistic looking woman. I don't know how you did it. It is perfect as far as I'm concerned. Thank you so much.
🍻