Test LoRa trained on 32 images in OneTrainer
Trigger - "The low angle image"
Works best with women
Model used to generate images V1 version is AcornIsSpinning FP8&NF4 (Full steps)
Model used to generate images V2 version is PixelWave dev03
Description
Trained on 1024px - better skin, updated captions
If you see horizontal/vertical lines - you can disable layers in LoRa (https://www.reddit.com/r/StableDiffusion/comments/1g7b76e/a_possible_fix_for_the_gridlines_effect_when/) or upscale without lora by <0.40
FAQ
Comments (1)
This is excellent, gives really good images from the low angle without interfering with other elements of the composition, really well done!
Details
Downloads
1,382
Platform
CivitAI
Platform Status
Available
Created
11/22/2024
Updated
6/12/2026
Deleted
-
Trigger Words:
The low angle image
from below'
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.