Update README.md
Browse files
README.md
CHANGED
@@ -16,7 +16,22 @@ tags:
|
|
16 |
# Kawai Diffusion v4-nightly (≧∇≦)ノ
|
17 |
See more in CivitAI : https://civitai.com/models/21138/kawai-diffusion-sd15
|
18 |
![](assets/preview.png)
|
|
|
|
|
19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
## Introduction:
|
21 |
- It's an AI art model for converting text to images, images to images, inpainting, and outpainting using Stable Diffusion.
|
22 |
- The AI art model is developed with a focus on the ability to draw anime characters relatively well through fine-tuning using Dreambooth.
|
@@ -66,6 +81,8 @@ Want to support me? Thank you, please help me make it better. ❤️
|
|
66 |
## Special Thank:
|
67 |
This wouldn't have happened if they hadn't made a breakthrough.
|
68 |
- [Runwayml](https://huggingface.co/runwayml/): Base model.
|
|
|
|
|
69 |
- [d8ahazard](https://github.com/d8ahazard/.sd_dreambooth_extension) : Dreambooth.
|
70 |
- [Automatic1111](https://github.com/AUTOMATIC1111/) : Web UI.
|
71 |
- [Mikubill](https://github.com/Mikubill/): Where my ideas started.
|
|
|
16 |
# Kawai Diffusion v4-nightly (≧∇≦)ノ
|
17 |
See more in CivitAI : https://civitai.com/models/21138/kawai-diffusion-sd15
|
18 |
![](assets/preview.png)
|
19 |
+
# What's new in Kawai v4-nightly:
|
20 |
+
Why did I name this version "nightly"? Just a random choice. Besides, Kawai is always trained late at night when everyone is already asleep.
|
21 |
|
22 |
+
The Nightly version allows you to use what we will do for the official LTS version.
|
23 |
+
|
24 |
+
- Feature:
|
25 |
+
|
26 |
+
- + The image quality has been significantly upgraded, thanks to a highly curated dataset.
|
27 |
+
|
28 |
+
- + Use VAE of stabilityai: [stabilityai/sd-vae-ft-mse-original · Hugging Face](https://huggingface.co/stabilityai/sd-vae-ft-mse-original)
|
29 |
+
|
30 |
+
- + True "kawaii"~~~
|
31 |
+
|
32 |
+
- + Img2Img is extremely powerful.
|
33 |
+
|
34 |
+
- + avatar: a new tag that allows you to create close-up images of a character's face. Suitable for use as a profile picture.
|
35 |
## Introduction:
|
36 |
- It's an AI art model for converting text to images, images to images, inpainting, and outpainting using Stable Diffusion.
|
37 |
- The AI art model is developed with a focus on the ability to draw anime characters relatively well through fine-tuning using Dreambooth.
|
|
|
81 |
## Special Thank:
|
82 |
This wouldn't have happened if they hadn't made a breakthrough.
|
83 |
- [Runwayml](https://huggingface.co/runwayml/): Base model.
|
84 |
+
- [CompVis](https://github.com/CompVis/): VAE Trainer.
|
85 |
+
- stabilityai: [stabilityai/sd-vae-ft-mse-original · Hugging Face](https://huggingface.co/stabilityai/sd-vae-ft-mse-original)
|
86 |
- [d8ahazard](https://github.com/d8ahazard/.sd_dreambooth_extension) : Dreambooth.
|
87 |
- [Automatic1111](https://github.com/AUTOMATIC1111/) : Web UI.
|
88 |
- [Mikubill](https://github.com/Mikubill/): Where my ideas started.
|