This LoRA is trained (on civitai) on the same data set as the SDXL/Pony otm gag. It adds knowledge for an over the mouth gag to FLUX 1.d. I have used 10 epochs during the creation and found out that epochs 6 and 7 are giving the best results. Still, it is my very first attempt to make a FLUX LoRA, so I don't know how well it will work. As all of us, we are still experimenting with it.It should add knowledge for terms like gagged and otm gag to FLUX.

Over the mouth gag - flux

The same model, 6th epoch (6 and 7 are providing better results IMHO, but feel free to try)

Files with SHA256: 54737a2212c0ba8a0c01a27e3a5735384baad97b53bfaf1143bc19d208edb1f3