This Z Image Turbo 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.
Trained on 19 in-game screenshots with hand written captions. Trained with ostris ai-toolkit using High Noise for Timestep Bias and Differential Guidance of 3.
No trigger word needed.
Based on the images used for my original Flux LORA https://civarchive.com/models/768707
Description
v1. 5000 steps. Trained on 19 in-game screenshots with hand written captions. Trained with ostris ai-toolkit using High Noise for Timestep Bias and Differential Guidance of 3.
Training only took an hour on my 4090. ZIT is crazy fast!
Details
Downloads
142
Platform
CivitAI
Platform Status
Available
Created
12/7/2025
Updated
12/25/2025
Deleted
-
Files
silenthill1 dataset.zip
Mirrors
CivitAI (1 mirrors)
SilentHill1_zimageturbo_v1.safetensors
Mirrors
CivitAI (1 mirrors)








