Ferdinand Clooney from Arknights
LoRAs are trained on all the images that I could find on the internet, official art and fan art
Preview generated without using ADetailer & Hires. fix
Recommended LoRA Strength (Weight): 0.7-0.8
v1.3 - The dataset has been slightly changed, new images and sprites with other emotions have been added. Trigger Words have been changed and Captions have been edited, this version should be better than other versions as much as possible (I will not claim that this version will be the last!)
v1.2 - slightly better looking hair and face than in v1.1 & v1.0 And added multicolored hair, spotted necktie tokens, but not sure if that helps (Most likely this will be the latest version).
v1.1 - is pretty much the same as v1.0 only softer not as sharp as v1.0 and trained a little differently
v1.0 - the first release, in comparison with other versions, looks worse
Description
Repeats: 10
Epoch: 20
Image Count: 16
Network Rank: 32
Network Alpha: 16
Batch size: 1
Clip skip: 1
Training steps: 3200
Resolution: 768x1024
Train Resolution: 1024x1024
Model: ponyDiffusionV6XL_v6
Captioning: WD14 Captioning + Edit (SmilingWolf/wd-v1-4-moat-tagger-v2)
Trigger Words: ferdinand-clooney, ferdinand-researcher, ferdinand-fugitive,
LoRA Weight: [0.7-0.8]
v1.2 slightly better looking hair and face than in v1.1 & v1.0 And added multicolored hair, spotted necktie tokens, but not sure if that helps (Most likely this will be the latest version).