Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
# For reference on model card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/modelcard.md?plain=1
|
3 |
+
# Doc / guide: https://huggingface.co/docs/hub/model-cards
|
4 |
+
{}
|
5 |
+
---
|
6 |
+
|
7 |
+
# Model Card for Model ID
|
8 |
+
|
9 |
+
New model focused on realistic images, based on 8Buff_Gen model
|
10 |
+
|
11 |
+
## Model Details
|
12 |
+
This model does require a good deal of prompt crafting, the trade off is you have a lot of control on the images you create. I would recommend finding a prompt that generates the style and quality to your liking and saving it as a style.
|
13 |
+
|
14 |
+
Use 512x512 size for most consistent results. Does Well with 512x768, 768x512, and 768x768.
|
15 |
+
|
16 |
+
Recommended sampling methods: DPM++ 2S a Karras, Euler a, DDIM, DPM2 a Karras, DPM++2S a,
|
17 |
+
|
18 |
+
Known issues:
|
19 |
+
|
20 |
+
Face distortion is still an issue but better than previous versions. Restore face and positive prompts address.
|
21 |
+
|
22 |
+
Tends to draw topless women about 10% of the time. This can be addressed by adding negative prompts nude and NSFw for image generation.
|
23 |
+
|
24 |
+
People tend to lean towards asian traits for eyes and hair. This can be addressed by adding nationality to the positive prompt. Nationalities that work with high consistency are: American, Indian, Nigerian, Japanese, and Ethiopian. Others such as European, Swiss, Mexican, work but not as consistently.
|
25 |
+
|
26 |
+
When using the Indian nationality prompt, it tends to give everyone a bindi. This can be addressed with adding (((bindi))) to your negative prompt.
|
27 |
+
|
28 |
+
Ages can be inconsistent sometimes with very young and very old people. Try positive prompts like
|
29 |
+
|
30 |
+
Improvements:
|
31 |
+
|
32 |
+
Requires less sampling steps to get quality images. Start at 30 and adjust from there.
|
33 |
+
|
34 |
+
Age consistency better with wider age range.
|
35 |
+
|
36 |
+
General purpose realistic model with a fine level of control. Merged with a lot of different models, but don’t remember all of them. Links below to those that were remembered or provided inspiration for direct training.
|
37 |
+
|
38 |
+
Trained on images and Mixed with:
|
39 |
+
|
40 |
+
8Buff_Gen - https://civitai.com/models/41928?modelVersionId=46665
|
41 |
+
|
42 |
+
Level4 - https://civitai.com/models/17449/level4
|
43 |
+
|
44 |
+
526Mix V1.4.5 - https://civitai.com/models/15022/526mix-v145
|
45 |
+
|
46 |
+
Great with deform and Automatic1111 Animator extension for GIFs or videos.
|
47 |
+
|
48 |
+
|
49 |
+
See more here for results: https://youtube.com/shorts/KwgcCbbQ35k?feature=share
|
50 |
+
|
51 |
+
Check out our How-tos here:
|
52 |
+
|
53 |
+
https://eightbuff.blogspot.com/p/my-text-to-img-journey-list-of-how-tos.html
|
54 |
+
|
55 |
+
Buy us a cup of coffee:
|
56 |
+
|
57 |
+
https://www.buymeacoffee.com/eightbuffalo
|
58 |
+
|
59 |
+
Follow up on Facebook:
|
60 |
+
|
61 |
+
https://facebook.com/EightBuffaloMediaGroup
|
62 |
+
|