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

Upside-Down Missionary - Concept

03
This model generates images featuring the "upside-down missionary" concept. It produces explicit content with specific triggers related to appearance and pose. The model's weight is 1.0.concept

LoRA · SD

specimen girl

33
This LORA model represents elements as specimens using the key token "Hyohon." The output is unstable and prone to artifacts, often generating arrangements of items, especially clothes, depending on the prompt. It can produce interesting results when used creatively, such as arranging specific objects.animeconceptgirls

LoRA · SD

Lafiel Abriel (Seikai no Monshou) LoRA

03
This LoRA model generates images of Lafiel Abriel from Seikai no Monshou. It was trained on NAI and works best at 0.8-1.0 strength, using "lafiel" as the trigger word. Adding "circlet, purple bodysuit, fingerless gloves, boots" helps achieve her signature appearance.animecharacterfemaleseikai no monshouwoman

LoRA · IL

Partially Submerged POV - Concept

13
This model generates images with a 'partially submerged POV' style. It focuses on scenes involving nude figures in or around water, often with elements of intimacy and a point-of-view perspective. The trigger word is "".concept

LoRA · Wan Video 2.2 I2V-A14B

Self Nipple Sucking Lora I2V - Wan2.2 - K3NK

4732
This LoRA (Wan2.2 I2V - K3NK) generates images of breast self-sucking. It was trained on a dataset of 319 clips across 5 different scenes at 211 resolution. It is designed for use with the WAN 2.2 I2V Kijai's Wrapper Workflow.actionbreastsnipple suckingnipples

LoRA · Wan Video 14B t2v

Detailz-Wan: Detail Enhancer for WAN Videos

12
Detailz-WAN enhances video generations, particularly with WAN models, by improving details in people, items, and backgrounds, as well as adding depth to colors and lighting. Trained on high-definition photographs, it works well with both text-to-video and image-to-video models, adding details and depth to the image without deviating too much from the input image. It is designed for the WAN T2V Model and strength of 0.75-1.0 is recommended.enhancestylevideowan

LoRA · SD

Taiwan Doll Likeness

62
This LORA generates images with a Taiwanese doll-like aesthetic, designed for use with photorealistic models. It works best with hires fix and higher resolutions, excelling at face variations but struggling with consistent replication and side views. This LORA is not open for commercial purposes.femalegirlsphotorealisticrealisticwomen

LoRA · IL

Convenient censoring | Selfie couple/lover | Concept version | Illustrious/NoobAI/Pony

02
This model generates improved images, particularly of censored selfie couples, with higher quality. It's optimized for resolutions of 832x1216 and a Lora force between 0.4 and 0.8. Monochrome and grayscale settings will produce a manga style.cellphonecheating (relationship)conceptconvenient censoringcoveringcovering crotchmirrornetorareselfie

LoRA · IL

Hug & Suck | Goofy Ai

02
Hug & Suck is a Lora model. Use it at 0.7-1 strength and ADetailer for faces, and consider img2img upscaling with 4x-ultra sharp. Check the provided images for prompt examples.animeconceptfellatiohug and suck

LoRA · IL

Caressing Testicles - Concept

02
This model generates images depicting caressing testicles. It includes associated visual elements such as genitalia, nudity, and suggestive positions. The model's weight is 1.0 and is activated using the trigger word "caressing testicles".concept

LoRA · Wan Video 2.2 I2V-A14B

Outsauce - Wan 2.2 I2V

2952
Outsauce Wan 2.2 I2V is a LoRA model that generates videos focused on penetration, specifically where the featured character penetrates someone off-screen in a doggystyle position. It is designed to work with males and transgender females and is capable of generating videos of varying lengths. The trigger word "outsauce" activates the LoRA, and structuring prompts with character names is recommended for better control over the scene.doggystyleinsertionsexstyle

LoRA · Wan Video 2.2 I2V-A14B

Wan 2.2 I2V Footjob

7812
Wan 2.2 I2V 14B is a model focused on generating images of footjobs, particularly toejobs. It comes in two versions, High and Low, which should be used together. While primarily trained on toejobs, it can also depict archjobs if feet are initially positioned accordingly.concept

LoRA · Wan Video 2.2 TI2V-5B

Futanari transformation | Wan 2.2 5B LoRa

492
LoRa To generate videos on the topic of "futanari transformation" There may be a problem because the model was trained on low-resolution video files.conceptfutanari

LoRA · Wan Video 14B t2v

Wan2.1_FusionX -The LoRa

02
Wan2.1_FusionX is a LoRa model that enhances image-to-video and text-to-video generation, improving prompt adherence, motion, and dynamics when used with Wan base models and SkyReels models. It offers a speed boost compared to the full FusionX model, enabling faster video rendering while providing control over the "FusionX" strength. The LoRa includes features such as causal motion modeling, improved temporal alignment, cinematic smoothness, and custom LoRAs for enhanced texture and clarity.fusionxi2vphantomstylet2vvacewanwan2.1

LoRA · SD

Futanari Horse Penis

42
This LoRA model generates futanari with masculine features and allows for various poses and angles. It can create hermaphrodite boys and provides guidance on avoiding issues like finger generation on the penis. The model also offers tips for adding skin textures and improving anime-style details.animeconceptfutanarisemi-realistic

LoRA · Wan Video 2.2 I2V-A14B

Ultimate Pussy and Anus helper

22
This model helps generating realistic female genitalia. I it inteded to be versitile so it does not force any pose, position or movement. You can just use natural language in your prompt, for some concepts specific phases are required. You can download the clips and drag them into ComfyUI. You can also look at my Profile for more samples. Concepts it can do (Trigger Phrases in Brackets) Any angle and distance (almost?) any pose Female masturbation (She rubs her vagina with her hand, she inserts her finger into her vagina) Pussy spread (She spreads her vagina with two hands) Can be used with Genitaila in the initial frame or not No public hair / With pubic hair (use "Hairy Vagina" in you prompt) Ass spread (She spreads her ass) Asshole winking (Her anus contracts repeatedly) Training Trained on ~140 completely hand captioned 5 second clips Usage Tips Greatly benefits from higher resolutions For maximum quality use 6+ low steps (10+ total steps) !!! Use with the default low WAN checkpoint, using it with WAN merges degrades quality Lora Strenght low model 0.9-1 high model: 0.5-1

LoRA · IL

[Takota Konu] Daitan Challenge ([蛸田こぬ] 大胆チャレンジ) - Concept

02
The model generates images based on the "Takota Konu Daitan Challenge" concept. Key features include explicit content such as nudity, ahegao expressions, and suggestive poses. Trigger words can be used to influence appearance and themes.concept

LoRA · Wan Video 2.2 TI2V-5B

Breast Expansion

12
This LoRA model, trained for Wan Video 2.2-5B, is designed to generate breast expansion effects in images and video using the trigger word "Br34st 3xpan510n". It exhibits more consistent breast growth than previous versions, but struggles with generating nipples from scratch and handling loose clothing on the affected area. The model was trained locally and initial testing included image to video implementations animated and photomedia content.breast expansionconcepthunyuan video

LoRA · SD

DoubleDick

32
nudespornrealisticstyle

LoRA · Wan Video 2.2 I2V-A14B

Sexy Move - Wan2.2 I2V

3342
This motion Lora, designed for stylized adult characters, injects 5 seconds of sexually suggestive movements like hip sways and self-caressing. It focuses on female-centric motion, ignoring clothing and anatomy outside of the "ass" tag, and is intended to enhance existing base model motions with added zest. The LoRA is best used with spicy prompts and a starting image that informs the desired motion.eroticgrindinghipshotmotionposessexy

LoRA · IL

Multiple Views - Concept

02
This model generates images with multiple views of a subject. It focuses on character appearance and can produce explicit content. The trigger word is ``.concept

LoRA · Wan Video 14B i2v 480p

Ahegao Face Wan (i2v)

32
This LoRA model generates ahegao faces, typically including drool and tongue protrusion. It was trained on a small dataset of videos and images and may sometimes produce sharp, vampire-like teeth. Users can try mitigating this by adding "sharp teeth" and "vampire teeth" to the negative prompt.ahegaoahegao faceconceptwan

LoRA · Wan Video 14B t2v

Wan 2.1 - Erect Penis, Cock, Dick, Lora, I2V

02
My other penis Loras can't do a penis facing more toward the front, so here's one that can. Main keyword is "penis" and it's also useful to say "facing the camera" or "facing to the side" "slightly to the side" to get different angles.concepterecterectionnsfwpenis

LoRA · SD

Ultimate Ahegao

31
Ultimate Ahegao Lora generates images with ahegao expressions. It offers three versions (V1A, V1B, V1C) producing images ranging from base model style to high contrast realism. Key prompts include naughty face, rolling_eyes, cross-eyed, ahegao, open mouth, tongue out, with a strength of 0.25-0.75.ahegaoanimeloraposeposesrealistic
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.