Trigger words aren't strictly necessary, but if you want to use one try "(NovaFrog:1.0)", it could help in some cases Also made a Textual Inversion of NovaFrog This is a model is trained on Realistic Vision V1.2 using photos from NovaFrog. Image prompting is an art unto itself. Something I am no good at. My results are kind of hit or miss. Got some keepers though. Someone that knows what they are doing might get better stuff. I may train it some more and see if consistency improves at all.
Training made a .yaml file too, not sure if it's needed. I included it anyway ?♂️. |