sexual situations

physical violence

disturbing

male nudity

hanging

hate symbols

nazi party

revealing clothes

weapon violence

female swimwear or underwear

male swimwear or underwear

partial nudity

white supremacy

adult toys

graphic male nudity

illustrated explicit nudity

nudity

graphic violence or gore

graphic female nudity

corpses

wide hips

convenient censoring

peeing

oral invitation

emaciated bodies

exposed female nipple

blowjob

female nudity

sexual activity

sexual intent

undressed

male underwear

female swimwear

genitals

female underwear

thick thighs

breasts out

strapless leotard

vore

breast out

one breast out

huge breasts

gigantic breasts

huge butt

covered nipples

hair over breasts

no panties

sitting on face

anal

dildo riding

downblouse

oral

porn

futanari

hentai

nude

lingerie

sexy

nsfw

suggestive

child on child

bestiality

self injury

extremist

hate speech

diapers

urine

menstruation

scat

incest

bukkake

cumshot

cumdrip

cum in pussy

cum on face

cum on hair

cum on body

cum on tongue

cum on hands

cum in mouth

cum on pussy

eat_cum

fellatio

implied fellatio

after fellatio

triple fellatio

autofellatio

pov fellatio

fucked silly

presenting ass

Created for a bounty for photorealistic people generated by software. This is a model trained on realistic cgi images of people generated in 3d modeling sofware and AI and intended to produce semi-realistic 3d style images in pony. Pony's base 3d image generation is rather inconsistent and runs the whole gamut from nearly photorealistic to cartoonish, and this model should steer it more in the realism direction. If there is a demand, I will also train the model for other checkpoints. As usual, I reccomend you check each versions' notes to determione what it does

Model trained on various images generated using DAZ studio using Civitai's training service. It uses much of the same dataset as the pony version, though I might have removed a fewer lower quality images from its set. I overtrained it a bit and some of the extraneous concepts got baked into the model, so it might randomly distorted characters at higher weights, so unless you are trying to get those concepts, I recommend using a weight around 0.7. It seems to produce a pretty consistent semi-realistic 3d render style across multiple checkpoints, and it works very well with sentence style prompts.

Iray Daz Render (SDXL)

Trained off of a bunch of daz renders of varying quality, none with any postwork done. Mostly human subjects, so it should be good at generating people, but it can also do monsters and machines. Doesnt seem to like space-ships though. "mydaz" is enough to trigger the style, but if you want something better looking, I would ad "realistic" or "detailed" to improve the quality. Some of the random concepts baked into the model will also work in other styles if you omit "mydaz". I recommend playing around to see what you can get.

Raw Daz Render (Pony)

Base version of the model,trained on about 60 CG images from various sources using civitai's training service. Thanks to AI_Photos for recommending me settings that helped increase the model's consistency. This version was mostly trained on CG images of people,some with exaggerated proportions, so if you get weird looking characters I recommend using negativcs to get rid of them. If you want a more photorealistic style,I recommend you put all of pony's 2d sources (Eg "source_anime","source_pony",etc) in the negatives to get a better result. I would reccomend checking the example images for the positives and negatives used and see what you like.

ThisPersonDoesNotExist V1

Semi-Realistic 3D Render Style

Type	LoRA
Stats	83 270 57
Reviews	Positive (13)
Published	Nov 10, 2024
Base Model	SDXL 1.0
Training	Epochs: 10
Usage Tips	Clip Skip: 1 Strength: 0.7
Trigger Words	iray
Hash	AutoV2 99649CFCFF