Porn Factory AILogin

girls models for adult image and video generation

Compare 278 public adult AI checkpoints and LoRAs tagged girls. Review each model's versions, base-model family, trigger words, previews, and validated image or video workflows before you create.

278 models available

LoRA · Pony

Hentai Pussy inspection 2shot Pony / IL XL | まんこ検査（完全着衣立絵と全裸くぱぁの2ショット）

3194
This model, Hentai Pussy Inspection 2shot, generates images featuring a fully clothed anime-style girl alongside a nude version displaying explicit content. It aims to create stable two-shot images, with specific prompts recommended for clothed characters. Two versions, IL and Pony, are available, each trained with distinct learning materials.analanimegirlshentainsfwpornposesvaginawoman

LoRA · Pony

X-ray (vaginal, anal & oral) LoRa | PonyXL & Illustrious

4184
This model creates anime-style images depicting visible penetration (vaginal and oral) through bodies. It's designed for PonyXL & Illustrious models. It is NSFW content and should be used responsibly.animeconceptgirlswomanxray

LoRA · Pony

Ingrid (Taimanin Series, Hell Knight Ingrid) NoobAI-XL (Eps / V-pred) / PonyXL / SD 1.5

3183
Ingrid (Taimanin Series, Hell Knight Ingrid) is a character model for PonyXL, NoobAI-XL, and Stable Diffusion 1.5. It generates images of the character Ingrid in various outfits like Hell Knight, business woman, secretary, and bride. The model leverages V-pred and Epsilon techniques for improved image generation and has specific quality settings for optimal results.animeart stylecharacterfemalegame charactergirlshentaiingridlorapersonrealisticsexyvideo gamewoman

LoRA · SD

Yae Miko | Realistic Genshin LORA

14176
This model generates realistic images of Yae Miko from Genshin Impact, utilizing specific embeddings and VAE. It comes in mixed, standard, and full versions, each optimized for different levels of compatibility with other LORA models. Optimal weights for image generation vary depending on the version used.animecharactergame charactergenshin impactgirlslorapersonphotorealisticrealisticvideo gamewomanyae miko

LoRA · SD

【Character / Art Style】Fashion Girl

14170
The Fashion Girl model generates images of fashionable girls, trained on a dataset of realistic and photorealistic images. It includes LyCORIS and LoRA versions, and the trigger tag is 'fashi-girl' for some versions. The model is continuously updated to improve image quality and fix issues.animeconceptgirlsrealistic

LoRA · Pony

Misty: Tomboyish Mermaid | カスミ (Pokemon) [Pony & Illustrious & SD1.5]

2168
This model generates images of Misty from Pokemon, specializing in recreating famous card art. It can produce images based on prompts, including specific outfits like her bikini and poses from various cards.animecharactercutefemalegirlsillustriousillustrious pokemonillustriousxlnintendoorange hairpokegirlpokemonsexytomboywater pokemon trainerwoman

LoRA · Pony

Pussy sandwich

7165
This model generates images of sexual acts. It uses tags like "ass," "pussy," and "tribadism," and benefits from additional tags describing penis presence or positions. The model is optimized for Illustrious anime styles and requires quality prompts.conceptgirlshentaivaginayuri

LoRA · IL

Aosiai123 | Shiiro's Styles

2152
The Aosiai123 model, part of Shiiro's Styles, generates images in a traditional media style using the trigger words "traditional media,aosiai123_illu." It's optimized with samplers like DPM++ 2M Karras and Euler a, a CFG Scale between 1-7, and a Clip Skip of 2, with a weight range of 0.1 to 1.abstract paintinganimeartart styleartistartstylebackground platebackground ponybackgroundscartooncolordetailflowersgame charactergeometrygirlsloramalemanmangamanga artistnsfwpaintphotorealisticrealisticshapesshiirostylestyleswoman

LoRA · SD

Nami (One Piece) Pre and Post Timeskip LoRA

4151
This LoRA model depicts Nami from One Piece, offering separate versions for her pre and post-timeskip appearances. It focuses on accurately representing her main outfits and distinctive tattoo, with better results using specific weights for each version. The model is designed to be compatible with most booru-based models.animebeautifulcharactercutefemalegirlsnamione piecephotorealisticportraitsrealisticsexywomanwomen

LoRA · Pony

Vacuum Fellatio | Concept Lora

1146
This Lora model generates images of a "vacuum fellatio" concept. Activation tags include "(:>=, vacuum fellatio)" and "vacuum fellatio, :>=". It supports tags such as "oral," "deep penetration," and "tongue out" to influence the image generation.:>=animeconceptconceptsfellatiogirlsoralsexual activitysexual actsvacuum fellatiowoman

LoRA · Pony

Glossy Skin (Illustrious+Pony)

1145
This LoRA creates glossy skin with circular reflections in generated images. It has been updated to improve quality and avoid overtraining with a unique trigger word. The model is designed for use with specific sampling and scaling techniques to achieve optimal results.animeconceptgirlsglossy skinloraoiled skinshiny skinwoman

LoRA · SD

Perfect Thick Ass

0143
This LoRA model generates images focusing on "thick ass, booty pose, big fat ass" themes. It is compatible with AUTOMATIC1111 and other WebUIs and can be loaded with specific code implementations. The model works best with ControlNet.big assbigassbootybooty posebutt poseconceptgirlslarge assthick ass

LoRA · SD

Panty Pull / Clothes Pull | Clothing Lora 446

5143
This Lora model is trained on 446 images to generate images of clothing, especially panties, being pulled. Due to AI limitations, hands pulling the garments may not be consistently generated. Example prompts provided enable the creation of diverse scenarios featuring this action.animeclothinggirlspanties

LoRA · IL

LoRA / illustriousXL / Improved Spread Pussy

3141
This LoRA model generates images with a specific focus on uncensored female anatomy. It is designed to be compatible with other LoRAs and uses specific prompts related to genitalia. The model offers adjustable clothing control via prompt triggers.2d2d artanimeanime charactercharactersconceptcosplaycostumecutegirlshentailadyloramemensfwpornposesexywoman

LoRA · SD

Goth Girl LoRA

4136
This LoRA generates images of goth girls, trained on NovelAI with a heavily NSFW dataset. Use the keyword 'gothgirl' along with descriptive tags for goth style; a strength of 0.6 to 0.8 is recommended. It's best suited for non-realistic models.animebeautifulboobsbreastscutefemalegirlsgothicgothicgirlhentainakednudespornsexsexystylewomanwomen

LoRA · Pony

Pepe (Jonasan) Style| Goofy Ai

1135
This model, named Pepe (Jonasan) Style| Goofy Ai, is inspired by the artist Jonasan. It generates images in a specific style that can be achieved by using the lora at 0.7-1 and ADetailer for faces. It is recommended to use Img2img upscale with 4x-ultra sharp for best results.animegirlslorasexystylestyleswoman

LoRA · Pony

Doggystyle | All angles | Doggystyle sex

2130
Doggystyle | All angles | Doggystyle sex Trigger word: Palidoggy Suggest strength: 0.5-1 Front view: (frontal, from the front) Back view: (male POV, his POV) Side view: This is the most common, shouldn't need extra promptsanimeconceptdoggy styledoggystlegirlsheterolorapenetrationsexsexywoman

LoRA · Pony

Backless Panties - Pony | Goofy Ai

2129
This model generates images of backless panties. It is recommended to use a Lora weight of 1, Highres fix, and Adetailer for improved faces and resolution.animeanusbacklessclothinggirlspantieswoman

LoRA · IL

JAV HARD BDSM Generator (bondage wall ) Pony / IL XL | 拘束少女快楽拷問 アダルト ビデオ ジェネレーター

1128
This LORA model is designed to generate hentai anime based on JAV (Japanese Adult Video) content, specifically focusing on "bondage wall" scenes. It allows users to create images with a desired anime character within the learned BDSM context, but requires adjustments to negative prompts to avoid photorealistic outputs. The model captures restraints, character positions, and potentially explicit acts.animebdsmbondageconceptgirlsnsfworalpornsexvaginawoman

LoRA · Pony

Mavis Dracula PonyDiffusionXL/illustriousXL

2127
Mavis Dracula PonyDiffusionXL/illustriousXL is a LoRA model for generating images of Mavis Dracula from the "Hotel Transylvania" series. It's recommended to use prompts like "m_d, 1girl, vampire, black hair, blue eyes, goth makeup" with a weight of 0.6-0.8 and the Fantasy Blend XL or Pony Diffusion XL models.charactergame charactergirlssexyvideo gamewoman

LoRA · IL

mon (monmon2133) style | Illustrious [Commission]

1124
This model replicates the art style of MON (monmon2133). It was trained on WAI-IL-V7 and is best used with specific samplers, schedulers, and CFG values, ideally with hires fix enabled. Recommended Lora Weight 0.8-1.0.aestheticanimegirlshorrorstyletheme

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

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 · 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 278

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.