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 · Qwen

2000s Analog Core

110
The 2000s Analog Core model generates images with a lo-fi, early 2000s analog aesthetic reminiscent of VHS tapes and grainy digital cameras. Key features include realistic timestamps, improved face variety, and training on VHS, VHS-C, and Hi8 footage. It's triggered primarily by the "v8s" keyword.2000s90scrtoldstyletapevhsvintage

LoRA · SD

Usagi/Bunny Tsukino/Sailor Moon - Sailor Moon

010
This model generates images of Usagi/Bunny Tsukino, also known as Sailor Moon, from the Sailor Moon anime. It was trained on the NAI model and works best with a preview strength of 0.6-1.0.characterfemalesailor moon

LoRA · IL

Fellatio Gesture - Concept

010
This model generates images depicting a fellatio gesture. It uses trigger words related to oral sex, suggestive poses, and facial expressions. Key features include depictions of an open mouth, tongue, and hand gestures associated with the act.concept

LoRA · IL

Oppai Challenge - Concept

010
This model generates images featuring a specific artistic style denoted as "Oppai Challenge." It focuses on depictions of characters presenting or flashing their breasts, often in suggestive poses and clothing. The recommended weight for usage is 1.0, and trigger words include "oppai challenge," "presenting breasts," and other terms describing revealing attire and poses.concept

LoRA · IL

(ILXL) Reverse Fellatio / 逆さフェラ

010
This model generates images of reverse fellatio. Key trigger words include "reverse fellatio," "upside-down," or "head back," with optional prompts like "oral" or "irrumatio" to improve stability. Perspective is controlled by prompts like "upside-down" but avoid using "head back" with detailed facial prompts.conceptnsfwreverse fellatio

LoRA · IL

Standing 69 LoRa | Illustrious

010
The model, Standing 69 LoRa | Illustrious, is trained for anime-style image generation. It is optimized for Illustrious models and uses trigger words to guide the artistic style. It may not be suitable for realistic images.69concept

LoRA · IL

On Side Missionary - Concept - COMMISSION

010
This model generates images of the "missionary on side" sexual position. It focuses on heterosexual vaginal sex with specific characteristics like leg positioning, facial expressions, and setting details.concept

LoRA · IL

Panty Pull Undressing From Behind - Concept

010
This model generates images depicting a panty pull undressing scene from behind. It focuses on elements like bare shoulders, underwear, full body views, and specific details of the lower body and bedroom environment. Trigger words include "panty pull undressing from behind" and related terms to control the image generation.concept

LoRA · IL

Goblin Rider / Frog embrace position

010
strength:1frog embrace positiongoblinposes

LoRA · IL

Reach Around Handjob - Concept

09
The model generates images depicting a "reach-around handjob" scene. Key features include depictions of nudity, sexual acts, and associated physical characteristics like sweat and blushing. The model is triggered by specific keywords and a LORA weight.concept

LoRA · Wan Video 2.2 I2V-A14B

Perfect Fingering - Wan 2.2 I2V

29
This LoRA model, "Perfect Fingering," aims to improve the realistic depiction of fingering motions in AI-generated images. It focuses on creating visually appealing and anatomically plausible finger movements, and may produce squirting effects with appropriate prompting. Recommended weight is between 0.7 and 1.0.conceptsex

LoRA · SD

Elegant Dresses Collection

29
This model generates images of elegant dresses. It is a Lycoris model, best used with specific sampler and CFG scale settings, and a Lycoris weight of 0.5 for optimal results.clothingfemalegirlsphotorealisticrealisticsexywoman

LoRA · SD

Murky's - Self Breast Sucking LoRA

39
This LoRA generates images of self breast sucking. It was trained on 50 images and uses the trigger word "SuckingOwnBreastMS", along with keywords like "huge breasts" and "nipples." Image generation may occasionally produce artifacts like extra nipples.animeerotic posepornposeposes

LoRA · SD

Female masturbation

39
This model generates images of female masturbation. Use the trigger word "female masturbation," and consider adding tags like "covering crotch," "fingering," or "masturbation" to enhance the effect. Note that issues with hand generation may be present.animeconceptscrotchfemalefemale masturbationfingeringhentaimasturbationposeposesspread legsvagina

LoRA · IL

Breast Sucking Upright Straddle - Concept

09
This model generates images depicting a "breast sucking upright straddle" concept. Key features include depictions of sexual acts and related anatomical details, with trigger words to guide image generation.concept

LoRA · IL

Lap POV - Concept

09
The model generates images with a "lap POV," simulating a view from below, looking up at a character from a lap pillow perspective. Key features include elements like blush, nude figures, navels, and smiles.concept

LoRA · SD

chracter-yukino

29
This LoRA model generates images of the character Yukino, trained using images from Pixiv. It is designed to work well with the NAIFU model (3.9G) and AOM2, but may also perform well with other models like anything-4.5. Downloading the dataset is suggested for optimal prompting and results.animecharacterfemalesexywoman

LoRA · IL

Breast Sucking Fingering - Concept

19
This model generates images depicting sexual acts. It is triggered by specific keywords related to breast sucking, fingering, and nudity.concept

LoRA · SD

Murky's - Standing Spitroast LoRA

39
This LoRA model generates images of the "spitroast" concept. It was trained with 768x768 resolution images and produces standing or dangling side-view scenes involving group sex.animepornposessexspitroastwoman

LoRA · IL

All Fours From Below - Concept

19
This model generates images in an "all fours from below" perspective. It focuses on nude poses with specific features like exposed navel, nipples, and a doggystyle orientation, often including elements such as motion blur and sweat. The trigger words help to achieve the desired appearance.concept

LoRA · SD

CharTurnerBeta - Lora (EXPERIMENTAL)

59
CharTurnerBeta is an experimental Lora model designed to influence your style and character. It is recommended for use with caution and at low weights, and is intended to be used in conjunction with charTurnerV2. This is an early beta release with known issues.consistent charactertoolutility

LoRA · IL

from behind breast press top-down bottom-up

09
后视图，挤压胸，撅起屁股，权重推荐：1~1.3，如果喜欢的话请麻烦点击follow~ 非常感谢，祝玩得愉快 from behind breast press top-down bottom-up，recommended weight: 1~1.3. If you like it, please take the trouble to click follow~ many thanks, and have fun!girlsloraposeposessexywoman

LoRA · SD

(N)SFW Filter / Slider | Tool LoRA

19
This LoRA model acts as an NSFW filter and slider, adjusting image content based on weight values from -3 to 2. Negative weights make output SFW, while positive weights generate NSFW content, overriding the prompt's inherent suggestiveness or wholesomeness.filtertooltoolsweight

LoRA · SD

ORCS

29
ORCS is a LoRA model trained on over 700 orc images (anime and 3D) to generate detailed SFW and NSFW orc characters. It's designed to work with NAI lineage models and allows for adjustable effect using the "orczor" tag and other descriptive tags. It aims to easily replace male characters in generated images.animecharacterhorrormonsterorc
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.