sexual situations

physical violence

disturbing

male nudity

extremist

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

self injury

graphic violence or gore

corpses

wide hips

convenient censoring

oral invitation

exposed female nipple

blowjob

female nudity

undressed

male underwear

female swimwear

genitals

female underwear

thick thighs

breasts out

strapless leotard

vore

bestiality

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

graphic female nudity

peeing

emaciated bodies

sexual activity

sexual intent

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

presenting pussy

spreading legs

spread legs

swimwear

This LORA is intended to reproduce the style of in-game graphics of Silent Hill (1999) for the PlayStation. It was not trained on the FMV cut-scenes.Suggested prompt: "sh1ps1 screenshot of "I'd recommend a strength of 1.5 to 1.9 for decent in-game looking results.Trained using fluxgym.This is my first crack on publishing a LORA. Please be gentle 🤣

Trained on 20 images, similar to v1 with a few swapped out. For v2 I used JoyCaption to caption the screenshots with natural language. I then made tweaks to the captions to help train the LORA to preserve Silent Hill's atmosphere. I'd recommend a strength of 1.5 to 1.9 for decent in-game looking results.

v2.0

Trained on 19 images. Mostly picked from the in-game cut-scenes so that I could eliminate any HUD elements and not have the screenshots include any one particular subject. I used minimal natural language captioning that I typed myself. I'd like to work on training a v2 in the future with a better dataset and better captions. v1 is more of an experiment. I'd recommend a strength of 1.5 to 1.9 for decent in-game looking results. Trained using fluxgym with a learning rate of 1.5e-4. Using the article by CrasH for guidance.

v1.0

Silent Hill PS1 In-Game Style

Type	LoRA
Stats	139 127
Reviews	Positive (20)
Published	Sep 18, 2024
Base Model	Flux.1 D
Training	Steps: 4,000 Epochs: 20
Trigger Words	sh1ps1
Hash	AutoV2 76D2094CD2