Run Locally
skindetails_krea2_loraholic.safetensors
7 MB · SafeTensor · bf16
Downloads are served by Civitai. Some creators ask for a Civitai login first.
About This Model
KREA2 · LoRAs
by Loraholic - Skin Detail Slider - (Krea2)
Creator notes
Skin Detail Slider
Krea2:
Unlike the Z-image version that worked best on the refiner pass alone, this works best at moderate values on all the generation passes you use in your workflow.
I had the same problem with showcasing this as i did with the Z-image version. A skin detailer should not be a huge flashy change that "pop" on sample images, it should change the "feel" of the image.
Sidenote, this is not the same as my at all. This will not fix the 3D/CGI look that some times happend that my Realistic slider was made to fix.
Recommended range: 0.5 to 3.5
Z-Image:
This is in my opinion, my best LoRA to date, but it has been the hardest to showcase.
I gave up making impressive sample images for this one. A skin detailer that's doing its job is invisible, you don't see it, you feel it. People stop looking like plastic and start looking alive, and the image itself doesn't change.
Don't trust my samples. Take a portrait or any image you've already generated, run it with and without and judge for yourself.
Use it on the first pass, the refiner pass, or both.
First pass only = more natural expression in the face.
Refiner only = pure skin texture on top of your existing image.
Both = fully alive, from expression down to the pores.
Recommended range: 0.5 to 4.0
Recommended range on refiner/detailer: 2.0 to 8.0
Moody and other checkpoints that makes faces a bit smooth can take higher values
One last thing to keep in mind, if you only use 1 pass when generating images, dont go to hard, or your pretty girls can turn, you know, a little to real. XD
All my work is 100% free. The only thing that really matters to me is people leaving a review, good or bad. I will NEVER remove or block anyone for leaving a bad review, so don't be afraid to give me criticism.
I'm stating this because when a creator blocks someone for leaving a bad review, that person can no longer see any of the creator's models, essentially muting their critics. This is something I'm highly critical of, and I believe it's a fundamental quality problem on this site.
SHA256
9C8537C435FADE7E2251C817D5D66512AF0203A1496161A4D9E65AF05874148B
ModelTrends.ai Model ID
#26673
Flag content
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
euler
Steps
12
Guidance
2
Seed
987750876
Size
832x1248
Weight used
0.8
Prompt
A zidiusArt
Negative prompt
deformed, bad anatomy, poorly drawn face, extra limbs, dull colors, blurry, muted lighting, text, watermark.
Settings
Sampler
Euler
Steps
6
Guidance
3
Seed
1112749565377461
Size
832x1216
Prompt
(real raw analog photo), (sharp focus, bokeh). real amateur photograph. amateur camera shot. casual snapshot. A candid 35mm film frontal portrait of a young woman around twenty years old with smooth, unlined skin and bright eyes, standing directly facing the camera in a supermarket aisle under harsh fluorescent lighting. She has dark chestnut brown hair styled in a sharp, chin-length bob cut with blunt ends that frame her jawline symmetrically. Her face features subtle metallic jewelry: a small silver stud piercing through her left eyebrow and another matching stud visible at her navel where the short crop top ends. She wears a white cotton crop top with a wide, asymmetrical neckline exposing her left shoulder and collarbone while keeping the chest fully covered by opaque fabric. Bold, slightly cracked black screen-printed text reading "I'll take it off for buzzz" is perfectly centered on the front, oriented flat relative to the camera lens with no perspective distortion. The fabric shows natural tension lines at the shoulders but remains completely matte and non-transparent. Her pose is grounded: both hands rest loosely on a shopping cart handle, weight evenly distributed, head tilted slightly with a playful smirk and direct eye contact. Background features blurred shelves of muted products melting into creamy bokeh circles, with no bright labels near the torso. Shot with sharp focus locked on her face, the crisp printed text, the detailed texture of the bob haircut, and the metallic glint of piercings. Authentic color film grain resembling Portra 400 stock, cool fluorescent highlights balancing warm skin tones, visible print cracks, cotton weave, and plastic shelf texture rendered without digital smoothing.
Settings
Steps
9
Guidance
1.3
Size
1664x928
Prompt
(vixen13_alt:1.2), 1 girl, solo, beautiful East Asian woman, fair skin, slender body, long straight wet black hair, wearing a sheer emerald green silk slip dress, oxblood red nail polish, holding a small crumpled paper sign with the handwritten text "I need 200 BUZZ" with delicate fingers, looking directly at the viewer, piercing gaze, a mix of submissive and predatory expressions, natural and seductive look, intense eye contact. Small glass table with a glass of red wine nearby, potted plants in dark pots amidst the shadows. Rooftop at night, heavy rain, rainy city skyline in the background, neon lights reflected in puddles, dark and evocative atmosphere, cinematic lighting, raindrops on skin and dress, shallow depth of field, city light bokeh effect, photorealistic, 8k, masterpiece.
Settings
Sampler
Euler
Steps
6
Guidance
3
Seed
897643556256158
Size
832x1216
Prompt
(real raw analog photo), (sharp focus, bokeh). real amateur photograph. amateur camera shot. casual snapshot. A dynamic wide-angle action shot captured with a 35mm lens from a low side angle, showing a man in his early thirties with curly brown hair, a faded elastic headband, and a thick cheesy 1970s mustache, sprinting down a suburban street. He is in mid-stride with both feet off the ground, body leaning forward, arms raised high above his head holding a rectangular piece of corrugated cardboard against the sky. The sign displays the handwritten English phrase "RUNNING FOR BUZZ" in bold uneven black marker letters, clearly legible against the overcast background. His grip is firm with fingers wrapped around the top edges of the cardboard, knuckles white from exertion. He wears a tight sweat-stained olive green t-shirt, matching short athletic shorts, and worn running shoes. Background features identical beige houses and empty asphalt streets under flat overcast daylight, with strong horizontal motion blur on the background elements to emphasize speed. Shot with shallow depth of field, sharp focus locked on the runner’s face, hands, and the distinct letterforms of the overhead sign. Authentic color film grain resembling Portra 400 stock, desaturated cool tones balancing warm skin highlights, visible fabric pilling, cardboard fibers, and imperfect letterforms rendered without digital smoothing.
Guides for this model
All guidesAI Models for Image, Video & Text Generation
skin details krea2 is a LoRA for the Krea2 family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Krea2 models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
Browse by Model Family
Text / LLM Models
411
Anima Models
509
Chroma Models
14
Flux Models
1,071
Flux 2 / Klein Models
1,237
MiniMax H3 Models
331
Hunyuan Models
197
Ideogram Models
20
Krea2 Models
674
Qwen Models
17
Qwen2 Models
36
Ltx2 Models
150
Stable Diffusion 1.5 Models
6,465
Stable Diffusion XL 1.0 Models
15,432
Zimage Models
804
Wan Models
919

Model
Trends
.ai
© ModelTrends.ai
|
Made in Japan
|
© 2026
