LoRAs are trained on images similar to Bara style, The dataset has so if you don't want to accidentally get such images add rating_safe in the prompt.
Not only guys! These styles can still create girls, furries, ponies. But it works a little better on guys
Preview generated without using ADetailer
Recommended LoRA Strength: 0.5-0.7
LoRAs have been trained on the Pony Diffusion V6 XL and I do not guarantee out good quality on other Pony Models
You can add photo, realistic, 3D, CGI, render, source filmmaker, to the Negative prompt to make the image less realistic, You can also add lineart, flat color, to the prompt to make the image flatter.
ㅤ
These styles are trained on just one image, so they may not always be stable, but most of the time the styles are as stable as possible for LoRA trained on just one image
These versions of the styles may have less contrast and saturation or a different color.
ㅤ
Styles are based partially or entirely on images created by the Neural Network (Artificial Intelligence)
Repeats: 100
Epoch: 20
Image Count: 1
Network Rank: 16
Network Alpha: 8
Batch size: 1
Clip skip: 1
Training steps: 2000
Resolution: 792x1152
Train Resolution: 1152x1152
Model: ponyDiffusionV6XL_v6
Captioning: WD14 Captioning (SmilingWolf/wd-swinv2-tagger-v3)
Recommended LoRA Strength: 0.5-0.7
1. 轉载模型僅供學習與交流分享，其版權及最終解释權归原作者。
2. 模型原作者如需認領模型，請通過官方渠道联系海藝AI工作人員進行認證。我們致力於保護每一位創作者的權益。 點擊去認領
