Porn Factory AILogin

AI porn models for adult image and video generation

Browse the building blocks behind the factory. Filter checkpoints and LoRAs by image or video support, base model, and tags, then open any version directly in the unified generator.

1487 models available

LoRA · SDXL

【SDXL】Retro_Illustration_Mix

01
Hi guys,this is Husky_AI, This is a oil and watercolor painting model based on SDXL merging: The triggering words are : oil and watercolor painting,2d game scene Preview images are all produced using SDXLbase recommend weight:0.7~1illustrationmixoilretrostylestyleswatercolor

LoRA · Wan Video 2.2 I2V-A14B

Ultimate BBC DeepThroat I2V Wan2.2 Video LoRa - K3NK

7381
The model is a LoRA for Wan2.2 I2V designed for generating videos featuring deepthroat scenes, primarily POV and side views. It was trained on a dataset of sex clips using automagic optimizer/learning rate and high/low noise models at different resolutions.actionbbc blowjobbbc deepthroatblowjobdeepthroatnsfwpov blowjobsideview blowjob

LoRA · Wan Video 2.2 I2V-A14B

Fuck Machine DeepThroat - Wan 2.2 K3NK

2941
actionblowjobdeepthroatdildodildo deepthroatfuck machinefuckmachinensfw

LoRA · SD

Morda (モルダ) witch Berserk Character LORA

01
Trainer on AnyLora Activation word: Morda Recommended weight : 0.7-0.9 Works well with anime models, tested on Anything 4.5 and AnyLora There was very little training data on the Internet, and no consensus on her hair/eye color, but the model turned out alright Use of " beautiful detailed eyes, " and "Detail Tweaker LoRA" LORA's is also recommended for good results I would appreciate any reviews, positive or negative! Enjoy :)animeanime girlberserkcharactercharactersmangawitch

LoRA · Other

Ahegao

141
The Ahegao model generates images of a specific facial expression. It requires specific trigger words and a weight of 0.4-0.9 (or 1.0 with some noise) due to the nature of the dataset. The model uses a manually curated dataset that includes various art styles, potentially including grayscale and monochrome images.animefemalehentainudespornsexsexystylewoman

LoRA · SD

Stella / Interstella 5555

01
This model generates images of Stella from Interstella 5555. Use a weight of 0.6-1 and the keyword stella5555. The standard outfit can be prompted with forehead_jewel, headband, two-tone_dress, and heart_cutout; high_ponytail can also be included.animecharacterfemale

LoRA · IL

Begging For It / Upside-Down Ahegao - Concept

01
This model generates images related to the "Begging For It / Upside-Down Ahegao" concept. It focuses on suggestive poses and expressions, including elements like blush, open mouth, and revealing clothing. Key triggers include "begging for it" and terms describing appearance.concept

LoRA · Wan Video 2.2 T2V-A14B

Ahegao Face - (Wan 2.2 and 2.1)

11
The Ahegao Face model (Wan 2.2 and 2.1) generates images of characters making the ahegao expression. It is triggered by the term "ahegao_face" and depicts crossed eyes and a protruding tongue. It requires both HIGH and LOW models to run effectively.concept

LoRA · Wan Video 14B i2v 480p

Face Sitting - WAN 2.1 I2V 480p

01
v0.2 Notes: Tweaked training procedure. Improved resolution and moves. Seems following prompts better and more stable in my case. Added some optional buzz begging for better training setups xD For generations I’m using 15–30 steps, a cfg of 7, and an 10 step shift for videos ranging from 32 to 64+ frames. Higher shift seems to be triggering the sitting effect faster in my case. The sweet spot for weight seems to be working 0.6-1.0, got some decent results with 0.65 but I have to say this one was tricky, sometimes it just creates body horror. Sometimes you just need to be lucky or have good conditioning starter frame... The only trigger word is: " facesitting "; the rest is up to your prompting. I recommend focusing on the action rather than the scene since this is I2V (at least that works for me). Here’s an example prompt I used in one of these generations—feel free to alter it heavily: facesitting, A woman from the camera. Her back is arched, and her ass is prominently displayed. She moves her hips slightly, causing her ass to shift then does the sitting movement. The camera angle is low, emphasizing her ass and the texture of her skin. I’m still experimenting the training process, so feel free to share your generations. This will help me gather more feedback to improve further. I didn’t have a large dataset to begin with and trained on relatively low settings if anyone’s curious. Since this is an NSFW-capable model, please use it responsibly. No real person faces were used in the training.

LoRA · Wan Video 2.2 T2V-A14B

Body Cumshot (Wan 2.2 / 2.1)

21
The Body Cumshot (Wan 2.2 / 2.1) model generates images of a woman's body covered in cum. It is triggered by the term "bodyshot" and depicts a scene with a large load of cum covering the woman's body. This version is designed to be more versatile in varying faces.concept

LoRA · Pony

Pillow Humping

01
This model generates images of a woman straddling and engaging in sexual acts with a pillow. The imagery includes grinding motions and simulated sexual intercourse.concept

LoRA · SD

M_Scene 卡通景景

31
M_Scene 卡通景景 generates cartoonish scenes in a toon or Pixar style with 3D elements. It's effective with the REV checkpoint and can create single buildings at smaller resolutions or sandbox environments at larger resolutions.3dbuildingscartoonfantasylandscapessencevideo game

LoRA · Wan Video 2.2 I2V-A14B

Full Nelson Sex I2V Wan2.2 Video LoRa - K3NK

4601
The Full Nelson Sex I2V Wan2.2 Video LoRa model generates videos based on image inputs, trained on a dataset of 76 short clips. It utilizes the WAN 2.2 T2V/I2V Kijai's Wrapper Workflow and is intended for responsible and lawful use only, focusing on explicit scenes.analfull nelsonfull nelson sexnsfwsex

LoRA · SD

Good Pussy

21
This model generates close-up images of female genitalia using the trigger word "mrrpss." It is trained on 71 images and works best for text-to-image applications with a weight of 1.0, and inpainting with a lower denoising strength.characterpornvagina

LoRA · Hunyuan Video

Sex Motion Helper (prototype)

31
Sex Motion Helper is a LoRA designed to enhance Hyvid hetero sex scenes by adding more motion and detail. It aims to improve existing sex LoRAs with features like a larger range of motions, better lighting, higher body detail, and crisper outputs. This LoRA is intended to be used in conjunction with other motion and character LoRAs using natural language prompts.conceptnsfwpornsex

LoRA · Hunyuan Video

HunyuanVideo Double Blowjob

21
This LoRA model generates images of two women performing oral sex from the viewer's perspective. It may not perfectly follow instructions, but produces decent results with a recommended strength of 0.7-1.0. Key phrases can help guide the image generation process.conceptoralsex

LoRA · SD

Najaran | Culdcept

01
Najaran From Culdcept (Old Manga) Very Limited dataset Strength 0.85+ if u want to generate the braid cloth earrings (Idk how to tag the earring part lol) u can adjust it depends on the checkpointsanimecharacterfemalegirlsmangasexywoman

LoRA · SD

Milfy City - Violet

01
This Lora model generates images resembling the character Violet from Milfy City. It is recommended to use a weight around 0.8, though lower weights down to 0.7 may be preferable for hyperrealistic models, triggered by the word "milfyviolet". The model was trained on Bismuthmixv3 but works well with others, particularly Real Dream.characterfemalegame characterwoman

LoRA · Wan Video 2.2 I2V-A14B

WAN 2.2 I2V - Teasing Sensual Blowjob

7721
This model generates images of women teasing or lightly stimulating a penis with their mouths, focusing on licking, kissing, and gentle sucking. It emphasizes eye contact and smiling but doesn't depict full fellatio. Users may need to combine it with other models for deeper penetration.actionblowjobfellatio

LoRA · Chroma

Lenovo UltraReal

11
Lenovo UltraReal is a LoRA model designed to enhance realism in generated images, particularly with Chroma models. It produces "live" images with dramatic lighting, shadows, and an amateurish aesthetic, and includes versions trained on Qwen, Wan, and a "Flux" dataset for unique effects. The model may require additional LoRAs for improved facial details.dramaticgirlsphotorealisticrealisticstyle

LoRA · IL

Panty Pull POV - Concept

01
This model generates images depicting a "panty pull" point-of-view. It features characters looking at the viewer with elements like blush, underwear, and various clothing items in an indoor bedroom setting, often on a bed. The images are designed with a focus on the character's appearance and are sometimes backlit with moonlight.concept

LoRA · Wan Video 2.2 I2V-A14B

Succubus Transform

01
She transform into a Succubus

LoRA · IL

condom on tongue/licking condom [ゴム舌置き] <ILL>

01
This is Lora licking a used condom or rolling it around on her tongue. It's also possible to lick it while it's still on.conceptcondomused condom

LoRA · SDXL

AbyCrimViso

01
AbyCrimViso is a Lora model designed to generate images with darker atmospheres. It excels at creating visuals depicting darkness, hooded figures, crimson colors, and shadowy scenes. The model can sometimes be unpredictable.darknessshadowstylestyles
Load more models

Showing 24 of 1487

How checkpoints and LoRAs fit together

Start with a checkpoint

A checkpoint controls the base rendering system. Its supported modes determine whether it can create images, edit an input, or produce video.

Match the base model

A LoRA must target the same base-model family as the checkpoint. Similar names do not guarantee compatible weights.

Check the validated workflows

A catalog version can be documented before it is generator-ready. Only versions with validated broker and artifact mappings can spend credits.