Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,134 @@
|
|
1 |
---
|
|
|
2 |
license: creativeml-openrail-m
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
---
|
3 |
license: creativeml-openrail-m
|
4 |
---
|
5 |
+
|
6 |
+
My pixiv: https://www.pixiv.net/en/users/92373922
|
7 |
+
|
8 |
+
# Doll Series by Kbr
|
9 |
+
|
10 |
+
The 'Doll-Series' is a set of LORA focused on realistic Asian faces, with impractical levels of beauty and aesthetics.
|
11 |
+
|
12 |
+
---
|
13 |
+
|
14 |
+
# Table of Contents
|
15 |
+
|
16 |
+
- [License](#license)
|
17 |
+
- [Disclaimer](#disclaimer)
|
18 |
+
- [Used Models](#use-models)
|
19 |
+
- [LORA Detail](#model-detail)
|
20 |
+
- [KoreanDollLikeness](#KoreanDollLikeness)
|
21 |
+
- [v10](#v10)
|
22 |
+
- [v15](#v15)
|
23 |
+
- [v20 (v2.0)](#v20)
|
24 |
+
- [JapaneseDollLikeness](#JapaneseDollLikeness)
|
25 |
+
- [v10]
|
26 |
+
- [TaiwanDollLikeness](#TaiwanDollLikeness)
|
27 |
+
- [v15]
|
28 |
+
- [FAQ](#faq)
|
29 |
+
|
30 |
+
|
31 |
+
|
32 |
+
---
|
33 |
+
# License
|
34 |
+
|
35 |
+
This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.
|
36 |
+
The CreativeML OpenRAIL License specifies:
|
37 |
+
|
38 |
+
1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
|
39 |
+
2. The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
|
40 |
+
3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)
|
41 |
+
(Full license here: https://huggingface.co/spaces/CompVis/stable-diffusion-license)
|
42 |
+
|
43 |
+
Additional Add-ons to license/notes
|
44 |
+
1. You shall take full responsibility for any creative work that uses this model
|
45 |
+
2. Refrain from using this model for malicious intent, harm, defamation, scam or political usages. It may impair and discourage the author from producing more works.
|
46 |
+
|
47 |
+
# Disclaimer
|
48 |
+
|
49 |
+
- Creation of SFW and NSFW images is user's decision, user has complete control over to generate NSFW content whether or not.
|
50 |
+
|
51 |
+
---
|
52 |
+
|
53 |
+
# Used Models
|
54 |
+
|
55 |
+
* Chilled_re_generic_v2
|
56 |
+
|
57 |
+
- https://github.com/wibus-wee/stable_diffusion_chilloutmix_ipynb
|
58 |
+
|
59 |
+
* chilloutmix_cilloutmixNi
|
60 |
+
|
61 |
+
- https://civitai.com/models/6424/chilloutmix
|
62 |
+
|
63 |
+
---
|
64 |
+
How to use:
|
65 |
+
- VAE: vae-ft-mse-840000-ema-pruned (For realistic models)
|
66 |
+
|
67 |
+
- Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)
|
68 |
+
- Steps: 20 to 35
|
69 |
+
- Clipskip: 1 or 2
|
70 |
+
- Upscaler : Latent (bicubic antialiased)
|
71 |
+
- CFG Scale : 5 to 9
|
72 |
+
- Denoise strength for img2img: 0.4 to 0.7
|
73 |
+
---
|
74 |
+
|
75 |
+
# LORA Detail
|
76 |
+
|
77 |
+
## KoreanDollLikeness
|
78 |
+
|
79 |
+
The first version that is widely used by many authors/AI artist/creators
|
80 |
+
|
81 |
+
### v10
|
82 |
+
|
83 |
+
- v10:
|
84 |
+
|
85 |
+
<img src="https://files.catbox.moe/r61ozj.png" width="" height="">
|
86 |
+
|
87 |
+
### v15
|
88 |
+
|
89 |
+
- v15:
|
90 |
+
|
91 |
+
<img src="https://files.catbox.moe/pgcfhc.png" widht="" height="">
|
92 |
+
|
93 |
+
### v20
|
94 |
+
|
95 |
+
- v20:
|
96 |
+
|
97 |
+
<img src="https://files.catbox.moe/thehrt.png" widht="" height="">
|
98 |
+
|
99 |
+
## JapaneseDollLikeness
|
100 |
+
|
101 |
+
The Japanese variant version, subjected for v15 in the future
|
102 |
+
|
103 |
+
### v10
|
104 |
+
|
105 |
+
- v10:
|
106 |
+
|
107 |
+
<img src="https://files.catbox.moe/cfypot.png" width="" height="">
|
108 |
+
|
109 |
+
## TaiwanLikeness
|
110 |
+
|
111 |
+
The Taiwan variant version, I have decided to discontinue v10, it is still out there somewhere in the internet, you may still find it.
|
112 |
+
|
113 |
+
### v15
|
114 |
+
|
115 |
+
- v15:
|
116 |
+
|
117 |
+
<img src="https://files.catbox.moe/5vr2z4.png" width="" height="">
|
118 |
+
|
119 |
+
|
120 |
+
---
|
121 |
+
# FAQ
|
122 |
+
- Q: Why can't I produce the same pictures as you?
|
123 |
+
- A: Sorry I don't share my full prompt, but you may ask me for advice on Twitter or Pixiv
|
124 |
+
|
125 |
+
- Q: What is the difference of each version upgrade?
|
126 |
+
- A: Version upgrade does not mean it will fix hands or legs, it is mainly difference of the face of the LORA, newer versions have wider range of faces.
|
127 |
+
|
128 |
+
- Q: Will you release all your other LORAs?
|
129 |
+
- A: Yes, maybe, but I would like to keep certain LORA to be exclusive to fans or supporters in the future through fanbox/ko-fi
|
130 |
+
|
131 |
+
- Q: Do you take requests or commissions on making custom LORAs?
|
132 |
+
- A: I might do community voting for requests, if you are somehow interested in a custom/exclusive LORA, you may contact my through Pixiv or Twitter for discussion
|
133 |
+
|
134 |
+
---
|