Teaches AI how to do underwater stuff.
The LoRa has mainly learned artist styles "@amanchu!", "@artbysaver" and "@rou22". But the model is flexible enough that it can work with other styles too. Just prompt for some style.
Scuba gear is a bit too complex perhaps and can often look bad. If it gets in the way negative prompt helps. Scuba gear was mainly captioned as "scuba", "diving regulator", "scuba tank", "flippers", "diving mask", "snorkel", "diving belt".
Older Illustrious version available at: https://civarchive.com/models/1317934/underwater-style-mix
Description
Based on Anima 1.0
Changes in training data tagging:
"@saver (artbysaver)" was changed to "@artbysaver"
"@rou (rou22)" was changed to "@rou22"
Trained on 848 images for 4000 steps with batch size of 2 and accumulation step 4. So 8 images per step.