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 · SD

MS Real - POV Blowjob LoRA

656
MS Real - POV Blowjob LoRA is a model trained for generating realistic images, though it also works with anime models. It was trained with 200 images at 768x768 resolution, and generates images of fellatio in a POV perspective. The suggested LoRA weight is between 0.8 and 1.pornposesreasex

LoRA · SD

Wet T-Shirt LORA

255
Fun LORA for getting a wet white shirt. Pretty self-explanatory what it does, but the ideal weight for this in my experiments has been .07-.09. Downside to this is that sometimes the wet shirt becomes liquid and I could not get it to do different colors than white for the shirt or smaller than H size boobs even though the training data has all these things.clothingphotorealistict-shirtwet

LoRA · SDXL

Touch of Realism [SDXL]

055
Touch of Realism V2 is a LoRA for SDXL that enhances image generation with a photographic look, introducing smoother lighting, better depth separation, and sharper detail. It adds realistic lens effects like soft focus, bokeh, and depth-of-field mimicking natural photography. It works best for portraits, landscapes, wildlife, and NSFW imagery, adding a touch of realism and cinematic softness.add realismphotorealismphotorealisticrealismstyle

LoRA · Pony

女児下着/girl's underwear(XL,ill,Pony)

155
This is the XL version that I created as a trial. I've just started using the XL version, so the quality is subtle. The strength of Lora is recommended to be 0.4~1.0.clothing

LoRA · IL

Fondled

255
The Fondled model produces images, with best results using a weight of 0.5-0.8 and CFG of 6-8. It currently has known issues with hands, which the next version aims to address.animebreast grabgirlsgrabbinggropegropingmanmultiple menmultiple_boysnsfwposessexywoman

LoRA · SDXL

BetterText.Redmond - Improving Text on SDXL Concept LORA

055
BetterText.Redmond is a concept LORA for SDXL 1.0 designed to improve text generation in images. It was fine-tuned on a large dataset and aims to produce images with coherent and accurate text. While not perfect, it enhances text generation capabilities within SDXL.stylestylestext

LoRA · Pony

Kafun STYLE -XL PONY | Illustrious XL- by YeiyeiArt

054
Kafun STYLE -XL PONY is a model designed to create art in the style of Marvel Rivals. Use the trigger word "KafunXLP".animecartoonfantasygirlsloraphotorealisticponyxlsexystylestylesvideo gamewomanxl

LoRA · IL

Evelyn Chevalier 伊芙琳・舒瓦利耶 | Zenless Zone Zero [IllustriousXL]

154
This model generates images of Evelyn Chevalier from Zenless Zone Zero. Key features include blonde hair, purple eyes, and detailed descriptions for her default outfit and an alternate version. The recommended weight is 1.0.animecharacterevelyn chevaliergame charactergirlswomanzenless zone zero

LoRA · IL

JESSIE_POKEMON_AnimeVER

154
JESSIE_POKEMON_AnimeVER is a model for generating images of Jessie from Pokemon in an anime style. It utilizes a weight of 0.7 and a specific prompt to create images of the character in her Team Rocket uniform.animeashboyscharactergirlsjamesjessiepikachupokepokemon

LoRA · Other

Virgin destroyer sweater

454
This model generates images of a "virgin destroyer sweater" paired with navel and thighhighs. It uses specific clothing and color tags for sweater and thighhigh appearances, including ribbed or aran-style sweaters. The model was trained with particular parameters for optimal performance, detailed in the technical specifications.animeconceptmeme attiresweater

LoRA · Wan Video 2.2 I2V-A14B

WAN 2.2 I2V - Combo Handjob Blowjob

854
This model generates images of a woman performing simultaneous hand and mouth stimulation. It works well at strength 1, supporting both POV and third-person perspectives, and is best suited for live-action images due to potential blurring on animated fingers. The model was trained to prioritize image quality over smaller file size.actionblowjobhandjobwan

LoRA · SD

2B (YoRHa No.2 Type B) / Nier: Automata

453
This model generates images of 2B from Nier: Automata, with weight 0.7 recommended. It includes outfits with and without her blindfold, which swaps between black blindfold/covered eyes and blue eyes. The model also enables switching between her dress and a leotard outfit, specifying details such as clothing cutouts, sleeves, and gloves.2bandroidcharactergamegame characterloranier automatayorha no.2 type b

LoRA · Pony

waist-grab missionary sex (pony, illustrious)

153
generate to missionary sex with boy grabbing her body (top-down bottom up). strength 0.8~0.6actionanimewoman

LoRA · SD

Nike Air Command Force

053
This model generates images of the iconic Nike Air Command Force high-top sneaker. It can be triggered with the keywords "commandforce" and specific colorways like "volt commandforce." The model works well on various styles, but users should be aware of potential color bleed across the image.basketballclothingshoessneakers

LoRA · SD

Sexy Attire | Dreamy Plunging Neckline Dress - by EDG

053
This model depicts a sexy attire featuring a dreamy plunging neckline dress. It is designed by EDG and available for image generation on Civitai. The generated images can be sold, but the model itself cannot be uploaded to other generation sites.clothingdressfashionfashion designsexy attirewoman

LoRA · IL

Fern [3 outfits] | Illustrious | Sousou no Frieren

053
This LoRA model generates images of Fern from Sousou no Frieren in three outfits, trained on anime screencaps. It is recommended to use ADetailer for detailed eyes and R-ESRGAN 4x+ Anime6B upscaler for high-resolution images. The model is compatible with various styles by adjusting weight and removing "anime screencap" from the prompt.animeanime charactercharacterfernfrieren beyond journe's endfrieren beyond journey's endfrieren the slayerfrieren: beyond journey's endgirlssousou no frierensousou_no_frierenwoman

LoRA · SD

Aqua (Konosuba) LoRA

252
This LoRA model generates images of Aqua from Konosuba. Use the trigger word "aqua (konosuba)" with a weight around 0.6, on anime-style models.animeaquacharacterfemalegirlsisekaikonosubasexywoman

LoRA · Flux.1 D

80s Fantasy Movie

152
This LoRA model captures the visual style of early 80s fantasy movies. It is triggered using the words ArsMovieStill and 80s Fantasy Movie Still, and it is available in multiple versions with different strengths and versatility. The model is trained on curated datasets of fantasy imagery and scenes.80sarsmjstylefantasymidjourney v6moviestillstyle

LoRA · IL

Mirko - My Hero Academia (IllustriousXL)

352
This model generates images of Mirko from My Hero Academia. It can create images of her base appearance with white hair, red eyes, and toned physique, as well as her hero outfit featuring a white leotard, fur collar, and purple thighhighs.animeboku no hero academiacharacterillustriousxlmirkomy hero academia

LoRA · IL

Starfire (Teen Titans) Pony + Illustrious

052
This model is a Starfire character from Teen Titans, designed to be highly accurate to the source material. It's an improved version compared to a previous Illustrious-based model, aiming for better parity with representations such as Raven.cartooncharactercomicdcstarfireteen titans

LoRA · SDXL

st. louis(azur lane)（圣路易斯）（碧蓝航线）

252
This model generates images of St. Louis from Azur Lane in various outfits. It features versions including her original look, a luxury handle dress, a blue and white pottery dress, spirits in the snow kimono, and an afternoon on the lido deck bikini. The model offers different versions like IL_v1.0, pony_v1.0, SDXL_1.0, and SD1.5_1.0, each with unique characteristics.animeazur lanecharactergamewoman

LoRA · SD

Underboob CropTop by Stable Yogi

052
The model generates images of characters wearing various underboob-revealing tops. It is triggered using keywords such as "Underboob CropTop," "Underboob t shirt," "Underboob bra," and "Underboob bikini," with a recommended weight of 0.8 to 1.0.actressanimeclothingfemalegirlslorapersonphotorealisticrealisticsexywoman

LoRA · Wan Video

Wan POV Doggy Style (i2v)

352
This model is trained for image-to-video generation of doggy style scenes, specifically optimized for vertical videos and using the trigger word "POVdog." Version 1.1 provides consistent motion for the doggy style position, with emphasis on ass movement and bounce, although it may have image artifacts due to the low resolution source video data. The model reportedly works well in conjunction with aipinups69's Wan 2.1 LoRA at a strength of 0.5.conceptdoggie stylepovvideowan

LoRA · NAI

X字マイクロビキニ/X micro bikini(SD,XL,illustrious,pony)

251
pony;The strength of Lora is recommended to be around 1.0. XL;The strength of Lora is recommended to be around 0.6.clothing
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.