Text-to-Image
Diffusers
stable-diffusion
lora
diaper
abdl
scat
omorashi
concept
Not-For-All-Audiences
Instructions to use mrpopsalot/aidl-omoaishi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use mrpopsalot/aidl-omoaishi with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ChenkinRF/ChenkinNoob-XL-v0.2-Rectified-Flow,Laxhar/noobai-XL-Vpred-1.0,OnomaAIResearch/Illustrious-XL-v2.0,neta-art/neta-lumina,CabalResearch/Mugen", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("mrpopsalot/aidl-omoaishi") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Commit ·
c8a1719
1
Parent(s): c9d8e26
Change the V16 changelog a bit
Browse files- Changelog.MD +2 -1
Changelog.MD
CHANGED
|
@@ -69,8 +69,9 @@
|
|
| 69 |
<h1>AIDL: OmoAIShi Changelog</h1>
|
| 70 |
<div class="section-div">
|
| 71 |
<h2>V16: Tag Restructure Edition</h2>
|
| 72 |
-
<p>Welcome to V16! Refined tags, another new customizable diaper element, and the new Neta Lumina
|
| 73 |
<p>Wetness indicators are now tagged to help stop the wetness indicators from appearing during pantypooping. You can also go and set up the wetness indicators to any color the model recognizes. The most trained colors are yellow and blue, which are common colors used for diaper wetness indicators, but if you use Neta Lumina, you can also set it to any color it knows.</p>
|
|
|
|
| 74 |
<ul>
|
| 75 |
<li>Added handmade natural language captions to <code>diaper</code> and <code>peed diaper</code> concepts images in the dataset.</li>
|
| 76 |
<ul>
|
|
|
|
| 69 |
<h1>AIDL: OmoAIShi Changelog</h1>
|
| 70 |
<div class="section-div">
|
| 71 |
<h2>V16: Tag Restructure Edition</h2>
|
| 72 |
+
<p>Welcome to V16! Refined tags, another new customizable diaper element, and the new Neta Lumina architecture brings AIDL to the world of Diffusion Transformers. While the Lumina variant is the standout variant of AIDL, the SDXL variants can still hold its weight well. With the Gemma 2 2B TE that Neta Lumina has, such concepts like <code>covered diaper</code> are possible without having any images of that concept, as well as the ability for natural language to be used.</p>
|
| 73 |
<p>Wetness indicators are now tagged to help stop the wetness indicators from appearing during pantypooping. You can also go and set up the wetness indicators to any color the model recognizes. The most trained colors are yellow and blue, which are common colors used for diaper wetness indicators, but if you use Neta Lumina, you can also set it to any color it knows.</p>
|
| 74 |
+
<p>Since TA went SFW only, and neither SeaArt and PixAI want to host the Lumina version, so this LoRA will only be on here, HF, and GitLab. I will try getting AIDL on alternative model sources such as LoRA Torrent and Diffusion Arc as well.</p>
|
| 75 |
<ul>
|
| 76 |
<li>Added handmade natural language captions to <code>diaper</code> and <code>peed diaper</code> concepts images in the dataset.</li>
|
| 77 |
<ul>
|