UPDATE: Consider checking out my new Interracial Gangbang (intrrclgngbng) checkpoint finetuned from the RealisticVision V6 B1 checkpoint for increased detail and a higher level of photorealism.


This is my first attempt at fine-tuning a LoRA intended to generate images depicting a general concept rather than images featuring a specific person. It does require some amount of fiddling in order to dial in settings that will generate good results relatively consistently. While I think there's still a lot of room for improvement, I don't know of any other models currently available which are capable of generating comparable images at a similar rate.

I used the clarity_2 SD1.5 checkpoint available on civitai as the base model and starting point for fine-tuning this lora.