|
|
--- |
|
|
license: unknown |
|
|
--- |
|
|
|
|
|
--- |
|
|
### Repository Description<br><br> |
|
|
Add models related to text-to-speech software characters. |
|
|
|
|
|
--- |
|
|
Contents currently available<br> |
|
|
1. kotonoha akane DB |
|
|
2. kotonoha akane HN(Monkey 1, 4, 1) |
|
|
<br><br> |
|
|
--- |
|
|
## Description of kotonoha akane DB<br> |
|
|
Usage Call token: akn |
|
|
<br><br> |
|
|
Only ".safetensors". Do you need ".ckpt" for extension or version reasons? If so, just do an |
|
|
``` |
|
|
A + A ( 1 + 0) |
|
|
``` |
|
|
merge from the "Checkpoint Merger" tag in Automatic1111 and make the save extension ".ckpt". |
|
|
This will give you a ".ckpt" with increased security. |
|
|
<br><br> |
|
|
This DB is in an overlearned state and is not very versatile. It is suitable for use with icons. |
|
|
<br><br> |
|
|
However, due to overlearning, a good "kotonoha sisters" lover may be able to identify datasets.<br> |
|
|
It is not suitable for Twitter icons or Pixiv submissions.<br> |
|
|
Perhaps they, the source of learning, will be very reluctant to have their illustrations captured by the AI.<br> |
|
|
 |
|
|
|
|
|
--- |
|
|
## Description of kotonoha akane HN<br><br> |
|
|
1. collect images of Kotonoha akane from Pixiv etc. |
|
|
2. put into "stable-diffusion-webui-wd14-tagger |
|
|
3. add "kotonoha_akane," at the beginning of "t2i<br> |
|
|
Easy, isn't it?<br> |
|
|
Recommend HN str < 1<br> |
|
|
<br> |
|
|
An even lower str is recommended for use with AOM2, etc. (Why don't you try starting at about 0.7?)<br> |
|
|
It worked for AOM2, so it should work for other mainstream merge models. Maybe.<br> |
|
|
<br><br> |
|
|
Oh, and oddly enough, the 5000step is corrupt. Should not be used.<br> |
|
|
<br> |
|
|
sample prompt |
|
|
``` |
|
|
parameters kotonoha_akane, 1girl, solo, pink hair, dress, thighhighs, long hair, open mouth, smile, blue thighhighs, v, white background, bangs, ribbon, wide sleeves, simple background, hair ribbon, black dress, hair tie, looking at viewer, :d, red eyes, blush, red ribbon, blunt bangs, sidelocks, long sleeves, cowboy shot, hair intakes, brown dress, pink eyes, hair ornament |
|
|
Negative prompt: lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts,signature, watermark, username, blurry, artist name |
|
|
Steps: 20, Sampler: Euler a, CFG scale: 10, Seed: 3300890004, Size: 512x768, Model hash: 5e5be6a7be, Hypernet: kotonoha_akane, Hypernet hash: ae9e4ada, Hypernet strength: 0.85 |
|
|
``` |
|
|
AOM2 |
|
|
 |
|
|
|
|
|
?.safetensors |
|
|
 |
|
|
|
|
|
 |
|
|
|
|
|
X/Y GRID |
|
|
 |
|
|
 |
|
|
 |
|
|
 |
