lazy-ti / README.md
ai-moroz's picture
Create README.md
a8b277b
|
raw
history blame
1.04 kB
---
tags:
- anime
---
### Lazy TI dump: Textual inversion embeddings host. Embeddings trained using `stable-textual-inversion-cafe Colab - Lazy Edition` with less than 100 images and most probably will not be perfect.
- Genshin Impact
1. <a href="https://huggingface.co/ai-moroz/lazy-ti/resolve/main/dump/diona-gi.pt">Diona</a>
<img src="https://huggingface.co/ai-moroz/lazy-ti/resolve/main/prev/diona.png" width="200">
<pre><b>diona-gi</b>, :3, solo, (standing), nature, birds, floating leaves, perfect fingers, bright, noon,(masterpiece:1.2), best quality, highres, original, perfect lighting, (extremely detailed CG:1.2),(8k:1.1)
Negative prompt: (worst quality, low quality:1.4), bad anatomy, blur
Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 8, Seed: 3253826251, Size: 512x768, Model hash: 791d67d4, Denoising strength: 0.6, Clip skip: 2, First pass size: 448x640</pre>
### uses
- Download and place the file in 'embeddings' folder
- Use the filename in positive prompt
#### preview models
- WarriorMama777/AbyssOrangeMix2