This LoRA model has been trained using Kohya_SS based on different dark fantasy images (with the same dataset as my Dark Fantasy Flux.1D LoRA) on Indecent checkpoint (itself based on Pony). This LoRA model allows to enhance dark fantasy artstyle on every realistic Pony checkpoints.

Increasing the weight of this LoRA will progressively increase the dark fantasy artstyle concept of the generated images until optimum weight values between 0.4 and 0.6. After, those values, images will become less and less realistic and more and more similar to digital paintings. See examples below :

Note : While the training dataset includes works created by several artists without direct authorization, this LoRA model's design ensures full compliance with copyright laws. This is achieved through a transformative training process that abstracts and generalizes some artistic styles rather than replicating specific copyrighted works. As a result, this LoRA model is able to generate innovative and original content inspired by several artistic influences, while remaining within the boundaries of legal and ethical standards.