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 · IL

Deep Penetration | LoRA

3124
This LoRA model generates explicit content, using "sex" as the trigger word. It is recommended to use a weight of 0.7-1 and the term "anus" instead of "penis" for certain positions.animeconceptdeep penetrationhentainsfwsex

LoRA · Flux.1 D

Painted World | Pony, SDXL, Flux

1122
This LoRA adds a painted style to images, emphasizing vivid colors and dynamic waves. It works well for 2D generations, landscapes, and backgrounds, and is available in Pony, SDXL, and Flux versions. Experiment with the LoRA weight to adjust the strength of the painted effect.animeartisticbackgroundbackgroundscolorfulpaintingscenery

LoRA · Pony

Niji oil painting drawing

1122
The Niji oil painting drawing model generates images in styles such as oil painting, drawing, hyper-realistic, and sketch, based on Nijijourney v6. Key trigger words like NiO1Pji and EOP can be used to influence the image style, along with options for anime/cartoon or semi-realism. Results are highly dependent on both positive and negative prompts.animecartoondrawinghyperrealisticloramannijioil paintingrealisticstylestyleswoman

LoRA · SD

Wetshirt - Wet T-Shirt LoRA

3121
This model is trained to make wet shirts without bras beneath them. Back in early April of 2023, this was the first LoRA I published! Technology and my own hardware and training skills have advanced quite a bit since then, so I'm happy to release version 2.0 to take advantage of those enhancements.clothingphotorealisticpornseethrushirtt-shirtwetwet t-shirtwettshirtwomanx-ray

LoRA · NAI

スライム姦/slime sex(SD,XL,ill,pony)

1121
The strength of Lora is recommended to be around 1.0. request loraconcept

LoRA · IL

Smooth Styles (Illustrious/NoobAI)

0120
Smooth Styles provides various artistic styles and effects when used with Illustrious and NoobAI models. These styles offer diverse visual outputs depending on the checkpoint and prompts used, encouraging experimentation. The model family includes a newly added "Lambda" style.animeanime characterscyberpunkfantasygame charactergirlshdillustriousrealisticscience fictionsemi-realisticsexystylestylesvideogame

LoRA · SD

Concept: Perfect Mouth

0120
This LoRA model is based on the concept of a "Perfect Mouth". It was created using Dreambooth, LyCORIS, and LORA techniques. The creator also takes suggestions and custom requests.conceptmouthsmileteeth

LoRA · SD

That's My Bush!

1120
That's My Bush! is a LORA model trained to add natural pubic hair to generated images. It works best with basic frontal poses and a weight of 0.2-0.4 and may cause unintended smiles. The model was trained on solo nude images using Realistic Vision 1.3 as the base.conceptfemalephotographyphotorealismpornportraitswoman

LoRA · SD

Bukkake LoRA

5120
This LoRA generates images with semen, specifically addressing limitations with existing prompts and models. It requires tuning for facial coverage and works best with models similar to AOM3, generating excessive cum with a simple "bukkake, cum" tag at a strength of 0.75. It offers marginal improvement for models that struggle with details like hands or genitalia.animecumcumshotfemalegirlshentainakednudespenispornsexual activitysexystylewoman

LoRA · Pony

Adjust details AND photorealism

0120
This model enhances images by adjusting details and photorealism. It can be used to add intricate details, improve lighting, sharpness, color grading, and overall composition to images; positive and negative strengths can be used to achieve either more or less realism and detail, respectively. It was trained for both illustrations and realistic photographs.conceptdetaildetailedenhanceenhancerphotorealismrealismrealistic

LoRA · SD

SaggyTitz

0119
V1.0 - Finally got this trained as a LoRa keyword: saggytitzareolamombodsaggysaggytitzstretchmarksstyletits

LoRA · IL

Eufoniuz (Anime Screencap)

0119
The Eufoniuz model generates artwork resembling anime screencaps. Using the tag "anime screencap" in the prompt is sufficient for V1, while V2 is a balanced version requiring less configuration. It's recommended to include "3D, realistic" in the negative prompt.animescreencapstyle

LoRA · SD

Ink scenery | 水墨山水

3119
先使用咒语“white background, scenery, ink, mountains, water, trees. Modify”。出了效果之后再更改、添加。 Start with white background, scenery, ink, mountains, water, trees. Modify after you successfully get the effect.backgroundinkmountainscenerytreewater

LoRA · SD

Mating Press (Experimental)

5118
This LoRA generates images of the "Mating Press" pose. It requires specific prompts like "matingpress, 1boy, penis, sex" to activate and was trained on 512x512 images. Be aware of potential issues like multiple testicles or disappearing limbs.animeconceptekuneheteromating pressmissionarynsfwsex발정

LoRA · IL

Xenomorph Queen - Aliens (1986) [IL/Pony]

1118
This model generates images of the Xenomorph Queen from the Alien franchise. It includes both Pony and Illustrious styles, with tags for features like sharp teeth, claws, and a tail. It also offers options to create a thicker-lipped or more feral version of the Queen.aliencharacterhorrormonsterscifiwomanxenomorphxenomorph queen

LoRA · IL

Matte Skin / Shiny Skin [Illustrious]

0117
Added v4.I recommend 0.5~1.5below is about v3. I recommend 0.5~1.0 v1 and v2 don't exist.skinsliderstyle

LoRA · IL

Annelid / Fellatio Monster

1117
This model generates images of "annelid" monster girls. Key features include eyeless creatures with prominent, dick-sucking lips, often with purple skin and variations allowing for plant-like, dragon-like, or latex-themed designs. The model uses specific keywords to trigger desired characteristics like body shape, mouth size, or thematic elements.characterfemalefictional creaturemonstersexywoman

LoRA · IL

Pressed Missionary (Feet on Chest) Multi-Views, Plus Cowgirl!

2116
This model generates images of the "pressed missionary" position with feet on the chest, and includes "cowgirl" position as a bonus. It was trained on three camera angles: "male pov," "from side," and "upside down shot." Specific tags and weight adjustments may be needed to achieve the desired results.feetfoot fetishposespositionssex

LoRA · Pony

ポールダンス/pole dancing

2115
The strength of Lora is recommended to be around 1.0.poses

LoRA · SDXL

Subtle Styles

1115
Subtle Styles is a collection of LoRAs designed to add versatile styles and details to images generated with the Big Love XL checkpoint and a few others. These LoRAs influence aspects like contrast, details, and sometimes colors, offering a range of effects from cuteness to artistic craziness. Each LoRA is trained on multiple images and recommended strengths vary depending on the desired style.fantasyscience fictionscifystyle

LoRA · SD

{DD} Tiptoes standing doggy

6115
This Lora model generates images of a "tiptoes standing doggy" pose. It offers versions with varying levels of control over limbs, character bodies, and position, and is best used with specific prompts and negative prompts to achieve desired results. The model can be used to depict the pose in realistic styles.dddoggystylehentaipornposessex from behindtiptoes

LoRA · SD

Raiden Shogun Genshin Impact | Character Lora 1200

2115
Lora trained on 1200 images of Raiden Shogun I release all my models for free, if you feel like supporting me and my work you can support me here and buy me a drink. Prompts: raidenshogundef, raidenshogunrnd Trained on NAI, gen'd on AOM2 hardanimecharactergenshin impactwoman

LoRA · Pony

Cum Through Clothing (Furry)

0114
Cum Through Clothing e621 tag, trained with Fluffyrock/Pony model. Best use with other furry base models. Prompt is cum through clothinganthroclotheingconceptcumcum through clothingfurrynsfwunderwear

LoRA · SD

【Character】alice (nikke)

7114
This LoRA model generates images of Alice from Goddess of Victory: NIKKE. It utilizes tag dropout to prevent overfitting and offset noise for better image lighting. The model is optimized for size and quality, and the NAI's VAE is recommended with weights between 0.6 and 0.8.animecharactergame charactergirls
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.