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

Chloe / クロエ - Kuroinu / 黒獣 - LoRA

06
This LoRA generates images of Chloe from the Kuroinu series using trigger prompts for her classic outfit and appearance. It is recommended to use a strength of 0.7-0.9, and the model can be combined with other LoRAs; additional costumes can be prompted using the reference images.animecharacterfemalegirlshentailorasexywoman

LoRA · Pony

restrained to table (illustrious, pony)

06
lora for girl restrained on the table.bdsmconcepthntai

LoRA · Wan Video 14B t2v

POV Handjob

16
The model generates images depicting a first-person ("POV") perspective of handjobs. The images focus on explicit sexual content involving naked individuals. It features scenarios with a man's viewpoint of a woman performing oral sex.concept

LoRA · SD

Petra Macneary (Fire Emblem: Three Houses) LoRA

06
This LoRA generates images of Petra Macneary from Fire Emblem: Three Houses. It includes both her academy and timeskip outfits, triggered by "petra_academy" and "petra_timeskip" respectively, with recommended prompts for her typical appearances.animecharacterfire emblemfire emblem three housesgame charactervideo gamewoman

LoRA · IL

Dildo Riding - Concept

06
This model generates images of characters riding dildos. It depicts scenes with specific features like spread legs, nudity, and expressions of struggle. The suggested trigger words help control the generated image's appearance.concept

LoRA · IL

(ILXL) Strengthen Grabbing Breast / 乳揉み

06
This model is a fine-tuned image generator that enhances the "grabbing breast" effect in images. It improves breast deformation, deep skin detail around fingers, and is more compatible with smaller sizes.conceptgrabbing breastnsfw

LoRA · Pony

present panties(XL,ill,pony)

06
The strength of Lora is recommended to be around 0.8. The fingers are not stable.concept

LoRA · IL

Kneeling Upright Sex From Behind - Concept

06
This model generates images depicting a kneeling upright sex from behind pose. Key features include depictions of nudity, sexual acts, and related characteristics like sweat, blushes, and motion lines. It is triggered by specific keywords.concept

LoRA · SD

Urushihara Satoshi/うるし原智志/漆原智志 《Langrisser》/《梦幻模拟战》 《フロントイノセント》/《Front Innocent》 - Artist Style

16
This model is trained on over 750 high-quality images in the style of artist Urushihara Satoshi. Version 3.0 offers improved character facial shapes, body movements, and detail. It is recommended to use a weight of 0.6-1.0 with animevae.pt VAE and scmix_NSM.ckpt model.front innocentlangrisserstyleurushihara satoshiうるし原智志梦幻模拟战漆原智志

LoRA · IL

handjob (boy lying)

06
lora for handjob.concepthandjobhentai

LoRA · IL

Hug and Suck - Concept

06
This model generates images depicting the "hug and suck" concept. It features explicit content, including fellatio and hugging. Appearance details such as "waist hug," "deepthroat," and other related triggers are used to influence image generation.concept

LoRA · Wan Video 2.2 I2V-A14B

wan2.2 NSFW Motion Enhancer

16
The wan2.2 NSFW Motion Enhancer model is a collection of LoRAs designed to enhance motion and details in NSFW image and video generation. It includes LoRAs for stomach bulges, penis generation (with improved structure), fluid dynamics (creampie/lactation), and reward/nudity enhancement. It also has a Reward LoRA which helps stabilize face consistency when using the "2d anime" prompt.base modelloransfwwan2.2

LoRA · SD

Harem dancer

36
This LoRA generates harem outfits, dancers, and face veils, sometimes producing see-through clothing. It focuses on creating face veils, and can also be used on male characters. Using "standing_on_one_leg" helps with anatomical correctness for dancing poses.animedancerface veilharem outfitsee-throughsexystyle

LoRA · IL

Multiple Ahegao Expressions Reference I Miscellaneous Concept I IllustriousXL and NoobAI

26
This model generates images of ahegao facial expressions, but results may vary and require rerolling for desired variance. It's best used with landscape resolutions and avoiding body/clothing tags for portrait images. It excels at creating images with tags like "multiple expressions," "ahegao," and "saliva."animecharacter_clothing_conceptexpressionspose_reference sheetstyle_virtual youtubervtuber

LoRA · SDXL

Envy Rough Oil XL 01

06
A rough painting LoRA for SDXL 1.0. Great for sci-fi, fantasy, landscapes, etc.digital paintingfantasygrungeimpressionismoil paintingpaintingsci-fiscience fictionsteampunkstyletexturewatercolor

LoRA · IL

Pearly lewd expressions Illustrious

06
This LoRA enhances lewd, erotic, and expressive facial features in generated characters, adding visual cues like flushed cheeks and suggestive mouth movements. It's designed for NSFW artwork to intensify arousal, teasing, and intimate tension. Use the trigger word "plexpressions".:oahegaoclenching teethconceptcrazy in pleasurecrazy laughdroolingfull face blushhands covering faceplexpressionspouting lipsvery embarrassedwavy mouth

LoRA · IL

(ILXL) Half Silhouette Sex / 半影姦

06
This model generates images of sexual content depicted as half silhouettes. The trigger word is "half silhouette sex", and scene-related prompts like "curtains" or "sliding doors" can influence the silhouette's appearance. It is recommended to use a weight of 1.conceptnsfw

LoRA · SDXL

o-ring dress ビッチ化 [XL, pony,illus]

05
This model generates images of characters wearing an o-ring dress and related accessories. It utilizes specified LoRA weights for stylistic elements and character consistency. Key features include gauntlets, thighhighs, a thong, and various revealing elements.bodyconclothingcorruptiono-ring dress

LoRA · SD

POV Missionary Anal + Creampie LoRA

75
This LoRA generates images of anal missionary scenes. Version 6 improves realism and the depiction of creampie/cum. It's recommended to use a LoRA strength of 0.5-0.8.actionanalmissionaryphotorealisticsexwoman

LoRA · IL

(ILXL) Hug From Behind-emphasize breasts / 後ろからハグ

05
This LoRA modifies the "hug from behind" pose to emphasize breasts, positioning hands around them. It can simulate a breast grab or choke hold using prompts like "strangling" or "grabbing another's breast." Recommended weight is 1.concepthugnsfw

LoRA · Flux.1 D

Defluxion

05
Defluxion is an image generation model, as showcased through provided images created with comfyUI in PNG format. The model's licensing conditions are aligned with the Flux Dev model.animeartcomicconceptsdrawingillustrationpaintingphotorealismstylestylist

LoRA · Flux.1 D

SamsungCam UltraReal

15
SamsungCam UltraReal is a LoRA designed to replicate the distinctive look of Samsung S-series smartphone photography. It produces images with vivid colors, realistic softness, HDR highlights, and less background blur. It is also compatible with NSFW models, and performs best at higher step counts.candiddigitalhighresiphonemodernsamsungstyle

LoRA · IL

group sex with a large number of boys

05
lora for a group sex with a large number of boys. strength ~0.8. It is better to use landscape form or 'vew fro above' to add more boys.concept

LoRA · SD

Murky's - Pronebone LoRA

45
This LoRA model enhances the "pronebone" sex concept in images. It was trained on over 400 images and can be prompted with terms like "sex from behind" or "lying, on stomach." For POV shots, use prompts like "ass, ass focus, pov, anus."animeerotic poseposessexsexy
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.