
LoRA is trained with videos of women taking off their clothes: taking off a shirt from over the head, unzipping a jumpsuit, revealing a trench coat, taking off pants and skirts—it has it all.
I’m very pleased to see Wan follow prompts so effortlessly. You can get what you want from 0.5 intensity to 1.0. However, the base model is quite bad at nudity, and the concept of genitalia is very strange. I recommend using a more general low-intensity nude lora (0.1 0.2) to compensate. I haven’t seen a 14B available yet, but there is a female model that can solve the issue.
I’ve added another post testing my lora using the I2V 480P model, and it seems to work well. Give it a try!
LoRA is trained with videos of women taking off their clothes: taking off a shirt from over the head, unzipping a jumpsuit, revealing a trench coat, taking off pants and skirts—it has it all.
I’m very pleased to see Wan follow prompts so effortlessly. You can get what you want from 0.5 intensity to 1.0. However, the base model is quite bad at nudity, and the concept of genitalia is very strange. I recommend using a more general low-intensity nude lora (0.1 0.2) to compensate. I haven’t seen a 14B available yet, but there is a female model that can solve the issue.
I’ve added another post testing my lora using the I2V 480P model, and it seems to work well. Give it a try!