This LoRA is designed specifically for the Hunyuan text-to-video generative model.

Trained on 42 videos, with manually extracted non-repetitive keyframes at 8 FPS and 33 frames per video, including 2D, 2.5D, and 3D styles. The style can be adjusted using prompts and weights, and it’s compatible with other LoRAs for blending.

The red section contains the primary reference costume appearance prompts, while the green section includes optional prompts.

prompt:

ex1: Anime-style digital drawing of a young woman with long, silver hair and blue eyes, wearing a white and blue outfit with a plunging neckline. She is performing oral sex on a man's erect penis. The man's penis is in the woman's mouth. The background is dark with a red pattern.

ex2: a young woman with long black hair and fair skin, performing oral sex on a man. The man is standing in front of her, with his erect penis in her mouth. The woman's eyes are closed, and she is looking up at him with a focused expression.

ex3: a young woman with long, straight, orange hair and purple eyes, wearing a black and white hooded cloak. She is performing oral sex on a man, who is partially visible, with his hands holding his erect penis. The background is a simple, blurred gradient. The woman's expression is one of concentration as she performs oral sex.

ex4: a young woman with long black hair and purple eyes, wearing a black top, performing oral sex on a large, erect penis. The penis is ejaculating onto her face and body. The background shows a dimly lit outdoor setting with a wooden railing and a pillar. The woman's expression is one of concentration as she performs the act.

!!!Attention

When mixing multiple LoRAs or setting LoRA weights very high, it's recommended to use only double blocks to improve the results. Here's how to do it: