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 · Pony

車内フェラ/car fellatio

117
The strength of Lora is recommended to be around 0.6.Since it has a large impact on the design, I plan to improve it.concept

LoRA · IL

Zero Suit Samus (Brawl)

017
This LoRA generates images of Zero Suit Samus based on her appearance in Super Smash Bros. Brawl. The model is strong in 3D style but can produce 2D results with prompting and negative prompts like "3D, realistic". Using checkpoints like AutismMix can also help achieve a less 3D style.charactermetroidnintendosamussuper smash brossuper smash bros brawlzero suit samus

LoRA · IL

Mouth Pull - POV - Illustrious

017
This model generates images featuring a first-person perspective of someone pulling another's mouth open. Key elements include open mouths, tongues, and hands pulling the mouth, and may include associated features like drool, tears, or other details. It responds best to prompts mentioning "mouth_pull," "mthpll," and related keywords.conceptcum in mouthcum in tongueillustriousmouth pulloralpovpov hands

LoRA · IL

NTR X-ray vision/NTR透視

017
This model is trained to create "x-ray vision" effects on anime characters, allowing users to explore a different side of them. It doesn't require specific trigger words, but using prompts like "see-through shirt" or specifying details like "covered nipples" can enhance the effect; ensure the character's outfit is defined in the prompt to avoid unwanted nudity. You can use methods like inpaint and controlnet to generate before-and-after comparisons.animationanimeconceptjapanesesee-throughwoman

LoRA · IL

Pony+Illustrious Disinterested Sex / Bored Female (Commission)

017
This model generates images depicting Pony+Illustrious Disinterested Sex/Bored Female in a commission style. It focuses on poses involving holding an object (initially a phone/book), and provides options to control the number of hands used. The model may exhibit unwanted purple/pink lighting which can be mitigated via negative prompts.conceptsph

LoRA · Wan Video 14B i2v 480p

Masturbation Cumshot for Wan I2V 14B

317
This LoRA model is designed for the Wan 14B I2V model to generate realistic cumshot scenes. It allows some control over where the cum lands using specific prompts and produces more realistic results on bodies and objects. The model was trained with improved datasets and captions at a higher resolution.conceptcumshotfutanarimasturbationpenistgirl

LoRA · SD

Caught cheating - NTR concept

416
This model generates images based on the "caught cheating" or NTR concept. It's an experimental LoRA that allows for complex scenes with multiple characters and poses depicting infidelity scenarios. The trigger word is "caught," and users should refer to example images for prompting guidance.animebackgroundhentainetoraresex

LoRA · SDXL

❤️SDXL Red Glitter❤️

016
This LoRA model trained on DALLE3 images adds a glittery, sparkly red effect to images. It is designed for use with SDXL FaeTastic 2.0 and works best at maximum strength, especially when prompting for themes like "angry" or "demonic".magicreindeersparklestylestyles

LoRA · IL

Human Onahole IL&NoobAI | Fantasy Sex Concept Collection

216
The model is a fantasy sex concept collection. It is titled "Human Onahole IL&NoobAI."cock sleeveconceptextreme size differencegianthentailiving sex toyonaholepornrough sex,sexsize differencestomach bulge,

LoRA · SD

girl and sundries

316
This model generates images of girls, replicating a specific art style from a Pixiv artwork. It works best with realistic models at 512x768 resolution, and high-definition upscaling is recommended. To generate SFW images, users can first create NSFW images and then refine them with ControlNet using the tile processor and appropriate prompts.femalensfwstyle

LoRA · IL

Standing Missionary

116
This model allows users to generate images, utilizing specific prompts and settings as indicated in the examples. Recommended settings include using a Lora strength of 0.7-1, employing ADetailer for face enhancement, and using img2img upscaling with 4x-ultra sharp. Users are encouraged to provide feedback and requests for future iterations.animeconceptsexstanding missionarywoman

LoRA · SD

Headpat POV | Concept LoRA

116
This LoRA model generates images from a headpat point-of-view. It uses a trigger word ("IncrsHeadpatPOV" for V2, "HeadpatPOV" for V1) and prompting such as "pov, headpat," with a recommended weight of 1. It is compatible with various styles, characters, and photorealistic models.animeconceptheadpovpov hands

LoRA · SD

Flashing Tits #4 (shirtsqueeze) SD1.5

016
This LoRA model for SD1.5 generates images of flashing tits via a clothing squeeze. It works best with clothing like dresses, shirts, tank tops, bikinis, or overalls and requires prompts including the clothing item and "tits".actionflashingtits

LoRA · Pony

Princess Carry Fellatio/Blowjob: Pony & Illustrious

016
The model generates images depicting a "Princess Carry Fellatio" scenario. It focuses on a female character carrying a male character in a sexually suggestive context. The model may have inconsistencies with finger count and requires specific prompt structuring for desired results.animegirlsloraposessexywoman

LoRA · IL

Hugging Doggystyle - Concept

116
Hugging Doggystyle is a concept model characterized by prompts including "hugging doggystyle," "hug," and related terms. It generates images featuring nudes, sex, and/or related depictions. The suggested weight for use is 1.0.concept

LoRA · IL

Penis Over One Eye - Concept

116
This model generates images featuring a penis over one eye. It utilizes specific trigger words related to appearance and body parts to create the desired imagery.concept

LoRA · NAI

セルフ挿入/guided penetration(ill,pony)

116
The recommended Lola strength is around 1.1.concept

LoRA · Flux.1 D

Niji Anime

116
The Niji Anime model generates images in PNG format and includes metadata/workflow. It uses specific settings like a weighting of 1.0, 25 steps, a CFG of 3.5, and the dpmpp_2m sampler with sgm_uniform scheduler, plus supports triggers for different styles.animefantasynijirealistic anime art stylesemirealisticstyle

LoRA · IL

group sex -acrobatic style-

015
lora for group sex with a girl suspended by boys. strength is 0.8~0.9.conceptgangbanggroup sexhentaispitroast

LoRA · IL

Arched back SEX with kissing! Doggy style v1.1 - IllustriousXL | NoobAI

015
This LORA generates anime scenes featuring a female character with an arched back kissing a partner. The key features include a dynamic arched back pose, detailed kissing interactions, and balanced compositions. It is triggered using the words "kiss, arched back, sex from behind" with a LORA strength of 1.arched backconceptdoggykissingsexsex from behind

LoRA · IL

Anal Object Insertion (Illustrious)

115
This model enhances images of anal object insertion. It was trained on Illustrious V0.1 datasets. Key tags include "analobjectinsertionILL," "anal object insertion," and "(object) in anus."analanal object insertionconceptmasturbationpenetration

LoRA · SD

Digital Oil Painting Style

015
oilpaintingstyle

LoRA · IL

Hand On Own Hip On Side - Poses

015
This model generates images of figures with a hand on their own hip, viewed from the side. It includes attributes like a blush, smile, navel, and collarbone, and can depict partially or fully nude figures. The trigger phrase is "hand on own hip on side" with a weight of 1.0.poses

LoRA · SD

Multiple penises

315
This LoRA generates images featuring multiple penises, either surrounding a character or in a gangbang scenario. Use trigger words like "multiple penises" and "multiple boys" along with tags like "surrounded" or "gangbang" to control the effect. Adjust the weight of the LoRA and tags to modify the output, and use negative prompts like "censored" to counter undesired censoring effects.conceptgangbanggroup sexhentaimultiple boysmultiple penises
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.