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

Bra Peek | Goofy Ai

251
Bra Peek is a model hosted on Tensor.art. Prompts are available in the images, use a Lora strength of 0.7-1, and ADetailer for faces with upscaling is recommended.animebraclothingfemalegirlslorapeekingsexywoman

LoRA · IL

Facesit / Facesitting POV (Illustrious)

251
This model is intended to enhanced images where the POV is an imminent facesit Trained on Illustrious V0.1 Tag: facesitpovILL, face sit pov • Helpful tags: impending facesit, imminent facesitting, facesitting pov, from below, standing over viewerconceptfacesittingnsfwpov

LoRA · IL

[ILLUST] Lillie | Pokémon Sun & Moon

051
This model generates images of Lillie from Pokémon Sun & Moon. It focuses on her appearance, including her hairstyles (twin braids, ponytail) and eye color, and can depict her in various outfits like her white dress or white shirt and skirt. Recommended weight is 0.7~1.0.animecharactergame characterlilliepokemonvideo gamewoman

LoRA · IL

N/SFW Filter / Slider | Tool / Concept LoRA IllustriousXL

051
This LoRA model filters N/SFW content in IllustriousXL, allowing users to adjust the level of explicit content by modifying the weight from -1.5 (SFW) to 1.5 (NSFW), regardless of the prompt's initial intent. It works with various styles, characters, and photorealistic models and doesn't require specific trigger words.conceptfiltertoolsweight

LoRA · SD

Gothic Sheer Leotard

151
This model generates images of a gothic sheer leotard. Avoid excessive LoRA usage for best results. The description includes a recipe for a Paralyzer cocktail.clothinggothgothicleotardsee-throughsheer

LoRA · SD

love hold

151
This model generates images of a "love hold" position, where one partner holds the other from the front with legs locked around their waist. It's typically generated with a male-on-top missionary position, and a separate LORA is recommended for optimal results. While trained primarily on heterosexual positions, lesbian versions might be possible by excluding male elements in the prompt.concept

LoRA · SD

Highleg Panties Thong

151
This is my first try making a LoRA so any feedback is appreciated. Im using the LoRA with a strength between 0.5 and 1.0. Trigger Word: highlegpanimeclothinghighleglingeriepantiesthongunderwear

LoRA · IL

Inkyo - Illustrious Custom Style

050
Inkyo creates images with a chaotic and energetic style reminiscent of vibrant cave paintings. It features bold colors, dynamic lines, and a sense of primal energy, resulting in visually striking and somewhat overwhelming artwork.artstylestyle

LoRA · IL

Ass Smack Lora - Slapping ass from behind - Updated to SDXL Illutsrious

050
This model generates images of ass slapping from behind, trained on relevant images for higher quality and accuracy. It is activated using specific prompts like "assslapsmack" and works with a weight of 1.0. The model allows for customization with prompts such as "ass," "thick thighs," and "from behind."conceptgirlsposesexyspanking

LoRA · Flux.1 D

OB半写实肖像画 Semi-realistic portrait painting

150
OB半写实肖像画 is a model for generating semi-realistic portrait paintings. Key features include enhanced facial details, reduced skin greasiness/redness, support for diverse age groups, and the addition of artistic and Chinese elements. It is designed primarily for European and American faces but canbe adapted with keywords for other ethnicities.artistic sensecolored pencil drawingportrait paintingsemi-realisticstyle

LoRA · Pony

[BSS] - Russian girls (Pony Realism)

049
The [BSS] - Russian girls (Pony Realism) model generates images in a Pony Realism style, encouraging imaginative and creative outputs beyond the provided preview images. It is copyright protected, prohibiting unauthorized copying or distribution.style

LoRA · SD

Vintage Gay Porn Magazine

049
A LoRA trained with 80s - 00s gay porn magazine page images. For better results, use danbooru tags style. Notes: • Best with: • CFG >= 7.0 • DPM++ SDE / dpmpp_sde / dpmpp_sde_gpu • Weight >= 0.55 • Worst with: • CFG < 5.5 • DPM++ 2M SDE / dpmpp_2m_sde / dpmpp_2m_sde_gpugaymagazinemalepornstyle

LoRA · SD

Sun and Shadow

248
Sun and Shadow is a noise offset trained Lora, based on the same dataset as Light and Shadow, but with distinct effects. It produces brighter colors and can depict sunlight, performing best with Anything and Silicon29 base models with a weight of 0.6-0.8.2dart styleartstylebeautifuldigital illustrationfantasymanstylevideo game

LoRA · IL

Dancer Lingerie

348
.clothesclothingdancerlingeriensfwnsfw clothes

LoRA · SD

ChilloutMixss3.0

348
AI工具:http://wogaosuni.cn (http://wogaosuni.cn) 0.2~0.6 Weight 1girl,

LoRA · IL

だいしゅきホールド/legs lock

247
The strength of Lora is recommended to be around 0.8.poses

LoRA · Pony

[SDXL] Darkness / 暗闇化

047
The SDXL Darkness LoRA generates dark images with bright light sources like lamps or stars, which is difficult to achieve through prompts alone. It is designed for 2D/2.5D illustrations and is compatible with Pony/Illustrious-based checkpoints. Adding prompts like "darkness" or adjusting the LoRA weight can modify the darkness level.concepteffectlora

LoRA · SD

Murky's - Spread Ass LoRA

747
Murky's Spread Ass LoRA generates images with a "spread ass" aesthetic. It was trained with 120 images and the trigger word is "SpreadAssMS"; the model's weight can be adjusted to control the intensity of the effect.hentaipornposes

LoRA · Pony

Dark Ghibli Fairytales

247
This LoRA combines Studio Ghibli visuals with dark fairytale themes, featuring watercolor backgrounds and cel-shaded characters in fantastical environments. It's inspired by artists like Hayao Miyazaki and incorporates MidJourney V7 to enhance versatility while preserving the original aesthetic.dark fairytalefairytaleghiblimidjourneystyle

LoRA · Pony

Pony+Illustrious Lying on Stomach / Lying on Belly / From Behind with sex

147
Comms always open, like and follow! Swimmer's position ass shot/ ass focus. Working on approximately a lot of LoRA's. If this one does well I will make a version with feet in the shot. New version can do sex reliably, Like and Follow, I am a vampire that feeds on attention!ass,buttconcept

LoRA · SD

Breast Massage SD1.5/SDXL/FLUX | olaz

447
This model generates breast massage scenes and works with both realistic and anime styles. It supports prompting via tags or descriptive text, with the latter often yielding better results, especially in the FLUX version. Face generation issues may occur but can be mitigated by adjusting the Lora's weight.massagensfwoiled skinphotorealisticposessexywoman

LoRA · Pony

フェラチオとオナニー/ fellatio and masturbation

247
The strength of Lora is recommended to be around 1.0. If you only use trigger words, it will be a normal, so please include other prompts as well.concept

LoRA · SD

Fern (3 Outfits) | Frieren: Beyond Journey's End

147
This model generates images of Fern from Frieren: Beyond Journey's End. It includes three outfits: her base outfit, winter outfit, and a dance dress, each with specific prompts for generation. The model's base appearance can be customized, and the recommended weight is around 1.0.animecharacterfemalefrieren beyond journey's endsousou no frierenwoman

LoRA · IL

POV Thighjob on Top / Lying on Top Thighjob - PONY/ILLUSTRIOUS

146
This LoRA model generates images of "POV Thighjob on Top / Lying on Top Thighjob" poses. The "Illustrious" version is recommended for dynamic pose adaptation and utilizes the trigger words: on 1boy, thigh sex, pussy, lying on viewer, on back, head on chest, POV, grinding, thighs together, penis. The "Pony" version fixes blurry faces and uses the trigger words: thigh sex, p0vth1ghj0b, lying on person, upside down, grinding, lying, on chest, male POV, legs together.conceptlying on toppovpov thighjobthighjob
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.