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

Double ended dildo LoRa | PonyXL & Illustrious

013
This LoRa model generates anime-style images featuring double-ended dildos. It's trained for and optimized with PonyXL & Illustrious models, using specific trigger words to activate the effect. It is not suitable for realistic images.animeconceptdildodouble digirlsobjectobject inssexywoman

LoRA · SD

no bra ; covered nipples

213
no bra, covered nipplesconceptcovered nipplesno bar

LoRA · Flux.1 D

Dark Side of Light

013
Dark Side of Light generates images, with showcased examples created using Fluxmania. It is recommended to use Clip L with the Fluxmania model, and all licensing conditions for the Flux Dev model apply.darkexperimental photofaded colorslow-keyphotogrphyphotorealimstyle

LoRA · IL

Dildo Anal Insertion - Object Insertion - Ass

013
Promptdildo_anal_insertion, dldnlnsrtn, 1girl, 1boy, anal object insertion, dildo, kneeling, ass, hips, top-down bottom-up, looking at viewer, dark-skinned male, faceless male, nude, ass focus, from behind, looking back, sex toy, object insertion, head out of frame,analbuttconceptdildoobject insertion

LoRA · IL

Loose Tank Top

213
This model generates images of loose tank tops. Use a LoRA strength of 0.7-1, ADetailer for faces, and consider upscaling with Img2img and 4x-ultra sharp. Check the example images for prompting ideas.animeclothingloose clothingloose shirtloose tank toptank topwoman

LoRA · IL

Cuddling Handjob/Nipple play | Concept

013
This model generates content related to handjobs and nipple play. It supports the inclusion of assertive female characters and femdom scenarios. Activation tags include "nipple_play" with supportive tags defining positions and character roles.conceptconcept loracuddling handjobfemdomforeplayhandjobnipple playnipple stimulationnipple tweaknsfwsexual activitysexywoman

LoRA · IL

Penis on Shoulder - Concept

113
This model generates images of a penis on a shoulder. It is triggered by phrases like "penis on shoulder" and related terms, and produces images containing nudity and sexually suggestive content.concept

LoRA · IL

Spitroast/Foot on head | Group sex/Sex position | Concept version | Illustrious/NoobAI/Pony

013
This model generates images featuring "Spitroast/Foot on head" and related themes, aiming for improved quality compared to previous versions. It is recommended to generate images at resolutions of 896x1152 or 1152x896 and use a Lora force between 0.6 and 0.8 for best results. The model can also generate images with a manga style.analconceptdefeatfellatiofoot on another's facefoot on headgoblingroup sexhumiliationimplied sexinterspeciesirrumatiomanon stomachprone bonerestrainedsexsmaller dominantspitroaststealth sex

LoRA · SD

nian/ニェン/年 (Arknights)

013
This model generates images of the Arknights character Nian, trained on a dataset of 1327 images. The LoRA allows users to recreate the character's features, with pruned character tags provided to improve stability. Recommended weights for the LoRA file are 0.5-0.85.arknightsbraidbreastscharactercolored skindragon taildragon_hornsearringshornslong hairmedium breastsmulticolored hairnianpointy earsponytailpurple eyesred hairred skinstreaked hairtailwhite hairニェン

LoRA · Flux.1 D

Realistic/Aesthetic Poses FluxDev

112
This LoRA model generates realistic and aesthetic poses, suitable for vintage or social media styles. It was trained with a limited number of steps and aims to provide poses not readily available elsewhere.aestheticaesthetic posedramatic lightdynamic poseflux.1flux1.dnightphotorealisticpinterestposesrealistic

LoRA · IL

Standing/Split stool sex | Size difference/Height difference | Concept version | IIllustrious/NoobAI/Pony

012
This model helps generate higher quality images of standing/split stool sex positions with size/height differences. It is a concept version trained on Illustrious/NoobAI/Pony and works best at resolutions of 896x1152 or 1152x896 with a Lora strength of 0.6 to 0.8. Use monochrome and grayscale to generate images with a manga style.conceptgoblinheight differencehumiliationinterspeciesnetoraresexsize differencesmaller dominantsplitstanding sexstanding splitstealth sexstooltaller femaleuncensored

LoRA · SD

Pearl Frush Pinup Style

012
This model embodies the pin-up style of Pearl Frush, a prominent American glamour artist known for her technically brilliant and often photographic-quality illustrations. Her work, popular in mid-century calendar art, features iconic depictions of women and a notable contribution to American advertising through her rendering of Little Debbie. Frush's artistry is celebrated for its detail and contributions to the pin-up genre.artpinupsstylestyles

LoRA · IL

standing_sex_machine

012
lora for sex machine with a girl standing. It can be used for multiple girls. strength ~ 0.8.bdsmconcepthentaisex machinestanding sex machine

LoRA · IL

NTR fingering/寝取られの手マン/背后的NTR

012
This LORA model generates images depicting NTR (Netorare) themes, originally intending to show a "fingering from behind" concept. It requires a trigger word and might produce different results, sometimes requiring multiple generations. Overfitting may occur, requiring weight adjustment.conceptfingerfingeringfrom belowntrupskirtwoman

LoRA · SD

Esthetic Futanari + Trap

712
This model generates aesthetically pleasing images of futanari and traps, prioritizing realism and retro styles. It offers variations with specific features such as chastity cages and condom depiction, and is trained on a large dataset with manual prompt settings. The model is ideally used with specific base models and negative prompts for optimal results, especially concerning anatomical accuracy and detail level through hires fix.aestheticanimebeautifulestheticfutafutanarihentaiotoko no kopenisstyletrapyaoi

LoRA · NAI

立ちバック/standing sex from behind

012
A strength of around 1.0 for Lora is recommended.concept

LoRA · IL

クロスビキニ/cross bikini

012
The strength of Lora is recommended to be around 1.2. I think it's interesting when used around 1.0.clothing

LoRA · SD

CleanMix_LORA

012
style

LoRA · IL

Stealth ass grab LoRa | Illustrious

012
This LoRa model is designed for generating anime-style images with a focus on "ass grab" poses. It works with Stable Diffusion and is optimized for Illustrious models. Use the trigger words to generate the desired effect.ass grabcheatingconceptnetorarestealth

LoRA · IL

Ass Worship | Goofy Ai

112
This model, "Ass Worship | Goofy Ai", appears to be a NSFW LoRA model. It can be used with prompts and tools like Adetailer for faces and img2img upscaling, with a suggested LoRA strength of 0.7-1.animeass worshipbuttconceptwoman

LoRA · IL

multiple male masturbation (illustrious,pony)

012
lora for cumshot by multiple male masturbation. strength~0.9.bukkakeconceptfacialhentaimale masturbation

LoRA · IL

Fatal Fury/The King of Fighters - Shiranui Mai l IllustriousXL (2 Outfits)

112
This model generates images of Shiranui Mai from Fatal Fury/The King of Fighters in two outfits. It features her original outfit and her City of the Wolves outfit, with associated prompts provided for each. Use a weight of 0.8 for optimal results.animecharacterfatal furyfatalfurygame characterkofmai shiranuithe king of fighterswoman

LoRA · IL

Breast Lift - Concept

012
This model generates images with a "breast lift" concept. It is triggered by terms like "breast lift," "grabbing own breast," and anatomical features, and aims for illustrative and intimate bedroom settings. The model uses a weight of 1.0 and incorporates specific camera angles and expressions.concept

LoRA · IL

POV Head Under Clothes | Extended Upskirt/Upshirt | PDXL/Illustrious

112
The POV Head Under Clothes LoRA generates images of characters from a point-of-view perspective, specifically under clothing. It supports various resolutions and benefits from specific trigger words in prompts, detailed in the showcase images. Version 2 works at 0.8-1.0 strength, while Version 1 is best at 0.9 strength, requiring "naked" in the prompt even when underwear is desired.conceptpovunder clothesupshirtupskirt
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.