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 · SDXL

ENF - Embarassed, Covering, Feminine (XL)

0254
This model, ENF - Embarassed, Covering, Feminine (XL), was trained on a hand-captioned dataset. It can be used at a 1.0 weight without quality degradation and has been tested upscaling at high resolutions.conceptcovering breastscovering crotchembarrassedembarrassed faceembarrassmentenf

LoRA · SD

cum on any

2253
This model improves semen quality, especially focusing on specific areas. It performs well with "cum on panties," "cum on pussy," and "cum on armpit" scenarios but is less effective with face shots, utilizing a strength of 1.0 though composition may collapse.

LoRA · SD

Hyperdetailed Colored Pencil

1253
This LoRA creates hyperdetailed colored pencil illustrations with intricate shading and fine textures. It's ideal for striking portraits featuring expressive details, natural beauty, and artistic sophistication. Keywords are ArsMJStyle and Colored pencil hyperdetailed realism.arsmjstylemidjourneymidjourney v6nijistyle

LoRA · SD

gangbang

3252
This Lora model generates images depicting gangbang scenarios. Use the trigger word "gangbang" and prompts such as "vaginal," "fellatio," "hand job," and "anal" to influence the image generation.analblow jobconcepthand jobmale_masturbationmultiple_boyssex pussy

LoRA · IL

Too Deep / Too Rough

2249
Pony and illustrious trained models for applying an expression during rough sex / penetration.ahegaoclench teethconceptpanicroughwide eyed

LoRA · SD

M legs | Pose Lora 800

4249
Lora trained on 800 images of m legs. You can probably do this just by prompting, you do not need this if you already know how to do this. Made on request. Trained on NAI gen'd on AOM2 nsfw.animeposeswoman

LoRA · IL

PowerPuffMixLora

1249
If you’ve seen my work or followed my journey, you probably already know what PowerPuffMixLora is all about. For those who don’t, I refined my portraits from the PowerPuffMix model and created a LoRA that enhances character aesthetics. The secret? Subtlety. Keeping the effect low is what makes the magic happen—at least, that’s how I do it. 😉style

LoRA · Wan Video 2.2 I2V-A14B

Pov_Blowjob

5246
This model generates images likely of explicit or sexual content. It utilizes Wan2.2 (High) v1.0 for denoising and Wan2.1 v1.0 for refinement.animebjblowjobgirlsloraphotorealisticposespovpov bjpov sexrealisticsexywoman

LoRA · IL

Takayanagi Katsuya style | Illustrious

1245
This model replicates the art style of Takayanagi Katsuya, a doujinshi artist, using a combination of poster and manga art. It allows users to generate images resembling Katsuya's work, and includes prompts for recreating his signature "paneled scenery" effect. It was trained on WAI-IL-v7.girlsillustriousxlmangastylestyleswomen

LoRA · SD

Onoff - On/Off LoRA

8245
The Onoff LoRA generates side-by-side images of a subject, with one version clothed and the other undressed. To use it, begin your prompt with "onoff" followed by a description of the subject and their clothing. Different versions of the LoRA are optimized for specific resolutions.conceptfemalenakednudephotorealisticsexyundress

LoRA · Hunyuan Video

Slime Girl Concept

1244
This model generates images of slime girls with diverse types and colors. It's a LoRA model best used with its trained base models as described in the accompanying "About This Version."animeconceptmonster girlslimeslime girl

LoRA · SD

Cumshot / Cum On Face

1244
💦💦actioncumcumshotnsfwporn

LoRA · IL

Thicc Slider Illustrious XL

3243
Thicc Slider Illustrious XL creates feminine figures with narrow shoulders, wide hips, and thick thighs on anime and realistic character bases. It uses a slider for fine-tuned adjustments to body proportions and enhances existing body shape tags without requiring a specific activation prompt.animebig assbimbobootyconceptcurvyfithuge assillustrioussexysliderthickthick thighsvenusvenusbodywide hipsxl

LoRA · SD

Busting out Massive Cleavage - fC 最大乳沟

1241
This model, named "Busting out Massive Cleavage - fC 最大乳沟," generates images focusing on large cleavage. The model might have applicability to anime-style images as well.clothingwoman

LoRA · IL

HoloEden Style (Illustrious)

1240
Use Remacri for better results. Holo Eden Holo_Eden (@HOlo_Eden) / X (https://x.com/holo_eden)barafurrygaymalemanmusclestyle

LoRA · IL

Dangling Legs LoRa | PonyXL & Illustrious

3240
This LoRa model generates anime-style images featuring girls with dangling legs, specifically in NSFW contexts with size differences. It's designed for use with PonyXL & Illustrious models and optimized for anime aesthetics.animeconceptgirlsposesexywoman

LoRA · Pony

Pony+Illustrious Futa on Female

3238
So Civit had a bug, when this hits 5k buzz I'll make it free. If you are reading this and it already reached the 5k threshold, message me. Like and Follow!conceptfutafutanari

LoRA · Flux.1 D

Hourglass Body Shape SD1.5/SDXL/PONY/FLUX | olaz

4237
This LoRA model helps generate images with hourglass body shapes, working for both realistic and anime styles. It addresses difficulties in prompting this specific body shape and is compatible with SD1.5, SDXL, PONY, and FLUX models.body shapesconceptnarrow waistsexywide hipswoman

LoRA · IL

Penis On Head

2237
This Lora model generates images featuring a penis on a head. It improves consistency and reduces variation compared to other models. Use the activation tags "PenisHeadER, penis" with a Lora weight of 0.8-1.0 for optimal results, and experiment with different positions and features like testicles or licking.concept

LoRA · SD

DPO (Direct Preference Optimization) LoRA for XL and 1.5 - OpenRail++

1237
This model provides DPO (Direct Preference Optimization) LoRA for Stable Diffusion XL and 1.5, finetuned based on human-chosen images to improve image quality and prompt adherence. These LoRA, licensed under OpenRail++, can be integrated with other fine-tuned Stable Diffusion models to enhance their ability to accurately reflect the given prompt.tool

LoRA · SD

Xeno Detailer - Magic Scene Detailer

0236
Xeno Detailer is a LoRa model designed to enhance images with intricate details and transform them into detailed artwork using various checkpoints. It adds object clutter, refines features, and can subtly adjust elements like hand poses and clothing styles based on the prompt. The model generally works best at weights up to 2.0, with lower weights adding general intricacy and higher weights trending toward opulence.art styleartstylesexystyletool lora

LoRA · Wan Video 2.2 I2V-A14B

Sex Smash Cut [WAN 2.2]

28236
Sex Smash Cut [WAN 2.2] is a pair of LoRAs trained on WAN 2.2 I2V 14B models to generate images of a man thrusting into a woman's vagina. They work best on wider aspect ratios and may occasionally produce body horror.conceptgonzojumpcutmissionarysexsmashcuttransition

LoRA · Pony

Pov Standing Doggy Style + Missionary (looking up) [remake]

4235
This model is a LoRA for generating POV sex positions, specifically standing doggy style and missionary. It features improved compatibility with other LoRAs and customization prompts for variations like breast press missionary. Core prompts are provided for default and alternate positions.animedoggystylehentailoramissionarynsfwposeposespovsexywoman

LoRA · IL

HL's Styles - WAI | ILLUSTRIOUS

1235
HL's Styles - WAI | ILLUSTRIOUS is a LoRA model offering a WAI/ILLUSTRIOUS version of styling options. Information about each specific LoRA within the model can be found in the "About this version" section.darkstylestyles
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.