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 · IL

口交 / Blowjob / フェラ | Face Fuck / 顔面ファック | Three-Panel Comic / 漫画 | Part 2

05
This model generates three-panel comic images, specifically focusing on explicit content. It is recommended to use a resolution of 640x960 with LoRA weight between 0.8 and 1, and a CFG scale of 4-7. Use "bjmcut" as the trigger word and consider adding other terms such as "blowjob" to the prompt.blowjobcomicconceptface fuckoralフェラフェラチオ漫画

LoRA · SD

Pussy Spread

35
This LORA model generates images of spread pussies, often with high detail. It struggles with accurately rendering the correct number of fingers. HiRez fix and NSFW-focused models are recommended for optimal results.anusnsfwnudespornpresenting pussyspread pussyvagina

LoRA · Pony

Threesome With After Sex IL&NAI&Pony | Shrekman Hentai Loras

05
This model generates images of threesome scenes. A key feature is its focus on depicting one participant as already having reached climax.after sexconceptthreesome

LoRA · SD

Uber Realistic Porn Merge (URPM) - LoRA Extract

65
This LoRA extract is based on the Uber Realistic Porn Merge (URPM) checkpoint and is designed for use with AUTOMATIC1111 WebUI. It can be used to generate NSFW images, and requires careful prompting with negative prompts to avoid unwanted content such as children. The model focuses on realistic depictions and various adult scenarios.base modelblendgirlshardcorelewdnudespersonphotographyphotorealisticpornportraitsrealisticsexsexyvaginawomanwomen

LoRA · SD

Murky's - Finger in Mouth LoRA

25
This LoRA model enhances the "finger in another's mouth" concept in images. It was trained on over 400 images and works well with related concepts like "cum on tongue," enabling the generation of both NSFW and SFW content.animeposes

LoRA · SDXL

バンドエイド/bandaid

05
The strength of Lora is recommended to be around 1.0.clothing

LoRA · IL

Dripping Pussy Juice From Behind - Concept - COMMISSION

15
This model generates images based on the concept "Dripping Pussy Juice From Behind." It uses specific trigger words to depict a solo female character with certain physical features and poses, often involving suggestive elements. The intended weight for use is 1.0.concept

LoRA · Wan Video 14B i2v 480p

Furry nsfw wan 2.1 14b img2vid (and nsfw motion in general)

35
This Lora is designed for the Wan 2.1 14b img2vid model and generates NSFW motion, including furry content, but can also be used with humans. It excels at maintaining character consistency in i2v generations and facilitates various positions and perspectives; prompting for desired motions and speeds is crucial. Using causvid lora is recommended for significant speedups.anthroconceptfurrymotionnsfw

LoRA · Wan Video 2.2 I2V-A14B

Big Cock Oral - Wan2.2 I2V

5435
This Lora model, trained on WAN 2.2 I2V 14B, generates images of oral or handjob actions, potentially featuring a large penis. It can function with or without a male figure present in the initial image.big cockconceptoral

LoRA · SD

Miss Heed | LoRa

05
Miss Heed from Villains I have tested several times and I think HeavenOrangeMix has the best results. On Weight I reccomend 0.6~0.7 Actually the LoRa has a hard time recreating the mask and eyes, but I still think is good enought. Any recomendation is welcomed.animecartoonnetworkcharactermiss heedvillainswoman

LoRA · SD

Garter belts

45
This model generates images of garter belts, often described as intricate lingerie items. It aims to depict garter belts as a symbol of love and harmony. Using provided prompts as a base is suggested for creating desired garter belt designs.fashionlingeriesexywoman

LoRA · IL

Slapping with Penis - Concept

05
This model generates images depicting the act of slapping with a penis. It is triggered by specific keywords including "slapping with penis," "hetero," "penis," and related terms. The model produces images with characteristics like erection, nudity, perspective views, and expressions of anger.concept

LoRA · SD

Murky's Lite - Standing Doggy Arm Grab LoRA

35
Murky's Lite - Standing Doggy Arm Grab LoRA is a less trained version of a full model designed to generate images of characters in a specific pose. It focuses on "sex from behind, standing, arm grab" positions and achieves best results at 512x resolution with hires fix, using LoRA weight between 0.8 and 1.animedoggystyleposessexwoman

LoRA · IL

Standing Sex Kiss - Concept

05
This model generates images of a "standing sex kiss" concept. Key features include depictions of sexual acts like kissing, intercourse and related physical attributes within a bedroom context. It is triggered using the provided keywords.concept

LoRA · SD

Sexy | mirror selfie

24
This Lora model generates phone-blocking selfies, working well with both anime and real-person styles. A weight of 0.6 is generally sufficient for good results. The model is being actively developed and updated.girlsphotorealisticselfiesexywoman

LoRA · IL

Imminent Penetration - Concept

04
The model generates images based on the concept "Imminent Penetration." It is triggered using specific keywords tied to appearance and suggestive poses. The model's weight is 1.0.concept

LoRA · IL

Breast Sucking Sex From Behind - Concept

04
This model generates images depicting a sexual act. It focuses on a scene of breast sucking during sex from behind. Keywords include specific body parts, positions, and elements like a bedroom setting.concept

LoRA · SD

Standing Full Body with Background Style LoRA (带背景立绘/背景付き立ち絵)

64
This LoRA is trained for generating standing full-body character illustrations with backgrounds, inspired by mobile game art. It supports both portrait and landscape orientations and offers control over background complexity and elements. Trigger words like "white background, full body" and prompt engineering can be used to customize the generated images.animeart styleartstyledigital illustrationfantasyfemalefull bodygirlsillustrationlandscapessimple backgroundstandingstylevideo game

LoRA · SDXL

Gambit XL

04
Gambit XL models the Marvel Comics character Gambit (Remy LeBeau), a mutant with the ability to manipulate kinetic energy. He is skilled in card throwing, hand-to-hand combat, and utilizes explosive projectiles charged with kinetic energy.animecharactercomiccomicsgame characterloramanmarvel comicsmarvel heropersonphotorealisticportraitsrealisticsexysuperherouncannyuncanny x-menvideo gamex-menx-men animated series

LoRA · SD

Milfy City - Liza

04
This LoRA model is trained to generate images of the character Liza from Milfy City, with a trigger word of "milfyliza" and a recommended weight of 0.8. It prioritizes resemblance to the game character over photorealism and may require extra prompting to achieve specific results, such as direct eye contact. It was trained on Bismuthmixv3 but is compatible with other models.characterfemalegame charactermilfwoman

LoRA · Wan Video 14B i2v 480p

Spanking for Wan I2V 14B

14
This Lora model is trained on the I2V model for generating spanking scenes. It uses videos from different angles and works with both hands and paddles. Controls are available through prompts to adjust action intensity and implementer.buttconceptspankspanking

LoRA · Wan Video

phut hon dance/大摆锤

14
Base on wan2.1 i2v model, 4 clips of phut hon dance video, example prompt： dabaichui，someone holds someone's head with both hands and twists someone's waist and hips left and right ，making dabaichui motion 本lora是开源lora，请不要用来为自己的商业用途进行引流以及其他获利操作，如进行此类操作请与本人联系action

LoRA · Wan Video 2.2 T2V-A14B

Dildo Masturbation - I2V & T2V

24
This model generates images and videos related to dildo masturbation. It focuses on scenes of a woman using a dildo vaginally and/or anally, including dual penetration.concept

LoRA · IL

Penis Shadow - Concept

04
The model generates images featuring a shadowy depiction of a penis, often within a suggestive or embarrassed context. It's triggered by keywords related to penis shadows, nudity, and specific angles or expressions.concept
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.