--- tags: - text-to-image - stable-diffusion - lora - diaper - abdl - scat - omorashi - concept - diffusers - not-for-all-audiences license: apache-2.0 base_model: - ChenkinRF/ChenkinNoob-XL-v0.2-Rectified-Flow - Laxhar/noobai-XL-Vpred-1.0 - OnomaAIResearch/Illustrious-XL-v2.0 - neta-art/neta-lumina - CabalResearch/Mugen ---

AIDL: OmoAIShi

V17: Pee-ka-poo Edition
For Anima, Mugen, Chenkin RF, NoobAI RF EQ-VAE, NoobAI RF FLUX.2 Vae, Illustrious 2

This model is designed for **NSFW** purposes, trained and tested on DRAWN CHARACTERS ONLY.

PUBLIC ALPHA. GENERATIONS DO NOT REPRESENT THE FINAL MODEL.

Static Badge Static Badge Static Badge Static Badge GitLab License

Overview

OmoAIshi is the most comprehensive multi-concept LoRA for diaper wearing and using, able to comprehend multiple concepts previously unfulfilled by base models and other LoRAs. LyCORIS LoCon is used instead of the regular LoRA format as it allows multiple concepts to be trained within the same model. No more girls in panties!


This is the first LoRA of the AIDL series. AIDL: The AI Diaper LyCORIS is intended to be a series of LoRAs that teaches models about diapers and other ABDL related concepts.

Showcase

Features

How to Use

Diapers

Use the general term: diaper, or in a diaper.

  • You can also specify the tape count using two tape diaper, or four tape diaper.
  • Prompting nothing else will get you a diaper with no designs, while print diaper will get you a printed diaper.
  • front-print diaper and back-print diaper can also be used to get you a print on the front or back. Use Danbooru's list of prints.
  • Add in wetness indicators by prompting yellow wetness indicator or blue wetness indicator.

Specific diaper styles (may or may not work, they have not been trained specifically):

  • pampers
  • huggies
  • luvs
  • abuniverse
  • merries
  • attento

Pull-Ups

Use pull-ups. Same prompting style as with diapers, but replace diaper with pull-ups.

Diaper Soiling

Peed-in diapers: Use the general peed diaper tag to get wet diapers. From there, you can also go and set how soiled the diaper should be. Without it, you might get unexpected leaks.

  1. wet diaper
  2. soaked diaper
  3. drenched diaper
  4. leaking diaper, pee on legs

Poopy diapers: Use the general poopy diaper tag to get messy diapers. From there, you can also go and set how soiled the diaper should be. Without it, you might get unexpected diaper blowouts.

  1. messy diaper
  2. packed diaper
  3. loaded diaper
  4. diaper blowout, poop on legs

smell can also be added to your prompt to give off that soiled smell effect.

If you want more soiled diapers, you can increase the weight of the tag to suit your needs. Tags work well with weights between 0.9–1.5.

Diaper Pull-Up

Want leaks and blowouts? Those can be done too

Pee Poop

Pantypooping

Use the term pooping self. You can also use the terms poopy underwear or poopy clothes to have either their underwear be soiled or their lower clothes (like their dress or pants) to be soiled.

Multiple underwear types supported!

For diapers please use the poopy diaper tag. Pooping self can also work but poopy diaper is the recommended tag for diaper pooping.

Open Diapers

Use the term open diaper to get open diapers.

  • Add in peed diaper and poopy diaper respectively if you want the diaper to be wet and messy.
  • As well, add your respective genitals such as pussy or penis and the uncensored tag

Covered Diapers

covered diaper is the main term used for diaper bulges. Works best on pants or shorts.

  • You can add diaper peek for waistband peeks. The color of the diaper may be chosen too.

Bias, Risks, and Limitations

The model currently struggles with the following concepts:


As well, there may be some limitations with this LoRA:

Increase the strength of the LoRA if needed to achieve your desired results.

I am in no way responsible for any content that the user creates with malicious intent while using any of these models. The user assumes complete responsibility for the misuse of these tools and any unethical content created with them. Follow all local, state, and federal laws as well as online rules regarding this content and generations with this LoRA. I am not responsible if you are arrested or banned from an online service for using this.

This LoRA violates CivitAI's terms of service (their TOS does not allow diapers nor scat). Do not upload generations from AIDL onto CivitAI.

Training Details

Over 1800 images have been used in the dataset. Images used are exclusively drawn 2D art.

No synthetic data has been used for the model. Tagging (Danbooru tag style) is all manually done, as automatic taggers do not recognize the concept in question well. Image captioning (NL support) uses a list of manual captioned images and (planned for V17) JoyCaption Beta 1 and Gemma 3 27B outputs with manual Tag grounding.

Training Procedure

Training configs are provided with the weights, but as a TL;DR:

NoobAI RF EQ-VAE

  • Program: Kohya sd-scripts Bluvoll fork
  • Optimizer: CAME
  • Scheduler: Constant
  • UNet LR: 2e-4
  • TE1 LR: 7e-5
  • TE2 LR: 3.5e-5
  • LoRA type: LoCon
  • Epochs: 16
  • Network Rank (Dim/Alpha): 64/1
  • Precision: Full BF16
  • Hardware: 1x RTX 3090 24GB

NoobAI RF FLUX.2 VAE

  • Program: Kohya sd-scripts Bluvoll fork
  • Optimizer: CAME
  • Scheduler: Constant
  • UNet LR: 3e-4
  • TE1 LR: 7e-5
  • TE2 LR: 3.5e-5
  • LoRA type: LoCon
  • Epochs: 16
  • Network Rank (Dim/Alpha): 64/1
  • Precision: Full BF16
  • Hardware: 1x RTX 3090 24GB

Illustrious 2

  • Program: Kohya sd-scripts Bluvoll fork
  • Optimizer: CAME
  • Scheduler: Constant
  • UNet LR: 2e-4
  • TE1 LR: 7e-5
  • TE2 LR: 3.5e-5
  • LoRA type: LoCon
  • Epochs: 10
  • Network Rank (Dim/Alpha): 64/1
  • Precision: Full BF16
  • Hardware: 1x RTX 3090 24GB

Chenkin RF

  • Program: Kohya sd-scripts Bluvoll fork
  • Optimizer: CAME
  • Scheduler: Constant
  • UNet LR: 1e-4
  • TE1 LR: 7e-5
  • TE2 LR: 3.5e-5
  • LoRA type: LoCon
  • Epochs: 10
  • Network Rank (Dim/Alpha): 64/1
  • Precision: Full BF16
  • Hardware: 1x RTX 3090 24GB

Anima

  • Program: LoRA Easy Training Scripts CLI
  • Optimizer: CAME
  • DiT LR: 4e-6
  • Scheduler: Constant
  • LoRA type: LoKr Full Matrix
  • Epochs: 20
  • Resolution: 512px
  • Network Rank (Dim/Alpha/Factor): 512/1/4
  • Precision: Full BF16
  • Hardware: 1x RTX 3090 24GB

Mugen

  • Program: Kohya sd-scripts Bluvoll fork
  • Optimizer: CAME
  • Scheduler: Constant
  • UNet LR: 1e-5
  • TE1 LR: 7e-5
  • TE2 LR: 3.5e-5
  • LoRA type: LoKr Full Matrix
  • Epochs: 16
  • Network Rank (Dim/Alpha/Factor): 512/1/4
  • Precision: Full BF16
  • Hardware: 1x RTX 3090 24GB

FAQ

License

This LoRA model is provided under the Apache License 2.0 license.

Special Thanks

Model Card Contact

Like my work? Support me by giving me a tip. This model will always be free, and open source. There will never be any paywalls, hidden fees, or any content locked behind a subscription.

Support
Regular options
Crypto
Chat

I do not have a SubscribeStar, as they do not seem to want to approve my account. I have never and will never open up a Patreon, as they are infamous for their ban waves and attacks against the NSFW community. Please support me on these alternative services.