Update README.md
Browse files
README.md
CHANGED
@@ -1,5 +1,76 @@
|
|
1 |
-
---
|
2 |
-
license: other
|
3 |
-
license_name: faipl-1.0-sd
|
4 |
-
license_link: https://freedevproject.org/faipl-1.0-sd/
|
5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
license_name: faipl-1.0-sd
|
4 |
+
license_link: https://freedevproject.org/faipl-1.0-sd/
|
5 |
+
pipeline_tag: text-to-image
|
6 |
+
base_model:
|
7 |
+
- RedRayz/illumina-xl-1.0
|
8 |
+
tags:
|
9 |
+
- stable-diffusion
|
10 |
+
- stable-diffusion-xl
|
11 |
+
---
|
12 |
+
|
13 |
+
# Illumina-XL-1.1
|
14 |
+
|
15 |
+
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/630e2d981ef92d4e37a1694e/zzWEQn_E9IN5HASpsqB2f.jpeg)
|
16 |
+
|
17 |
+
Modified Illustrious-XL-v0.1
|
18 |
+
|
19 |
+
Your can find sample images on [Civitai model page](https://civitai.com/models/832013)
|
20 |
+
|
21 |
+
## About v1.1
|
22 |
+
Slightly improved quality(stability) of background(scenery) and style
|
23 |
+
|
24 |
+
## Prompt Guidelines
|
25 |
+
Almost same as the base model
|
26 |
+
|
27 |
+
## Recommended Prompt
|
28 |
+
### standard
|
29 |
+
Positive: None(Works good without `masterpiece, best quality`)
|
30 |
+
|
31 |
+
Negative: `worst quality, low quality, bad quality, lowres, jpeg artifacts, unfinished`
|
32 |
+
|
33 |
+
Negative(more stability): `worst quality, low quality, bad quality, lowres, jpeg artifacts, unfinished, oldest, old, photoshop \(medium\)`
|
34 |
+
|
35 |
+
### masterpiece style
|
36 |
+
Positive: `newest, ai-generated`
|
37 |
+
|
38 |
+
Negative: `worst quality, bad quality, lowres, jpeg artifacts, unfinished, comic, 1koma, 2koma, 3koma, 4koma, oldest, old, photoshop \(medium\)`
|
39 |
+
|
40 |
+
## Recommended Settings
|
41 |
+
Steps: 14-28
|
42 |
+
|
43 |
+
Sampler: DPM++ 2M(dpmpp_2m)
|
44 |
+
|
45 |
+
Scheduler: Simple or Karras
|
46 |
+
|
47 |
+
Guidance Scale: 4-9
|
48 |
+
|
49 |
+
### Hires.fix
|
50 |
+
|
51 |
+
Hires upscaler: 4x-UltraSharp or Latent
|
52 |
+
|
53 |
+
Denoising strength: 0.4-0.5(0.6 for latent)
|
54 |
+
|
55 |
+
|
56 |
+
## Merge recipe(Weighted sum)
|
57 |
+
I made 6 Illustrious-based models and merged them.
|
58 |
+
|
59 |
+
*A-F: illustrious-based custom checkpoint
|
60 |
+
|
61 |
+
- Illumina-XL-1.0(Enhanced background) * 0.55 + A * 0.45 = tmp1
|
62 |
+
- tmp1 * 0.8 + B * 0.2 = tmp2
|
63 |
+
- tmp2 * 0.65 + C * 0.35 = tmp3
|
64 |
+
- tmp3 * 0.6 + D * 0.4 = tmp4
|
65 |
+
- tmp4 * 0.6 + F * 0.4 = tmp5
|
66 |
+
- tmp5 * 0.85 + E * 0.15 = Illumina-XL-1.1
|
67 |
+
|
68 |
+
## Training scripts:
|
69 |
+
[sd-scripts](https://github.com/kohya-ss/sd-scripts)
|
70 |
+
|
71 |
+
## Notice
|
72 |
+
This model is licensed under [Fair AI Public License 1.0-SD](https://freedevproject.org/faipl-1.0-sd/)
|
73 |
+
|
74 |
+
If you make modify this model, you must share both your changes and the original license.
|
75 |
+
|
76 |
+
You are prohibited from monetizing any close-sourced fine-tuned / merged model, which disallows the public from accessing the model's source code / weights and its usages.
|