Commit
·
d976c32
1
Parent(s):
d5e9bad
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,89 @@
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
---
|
4 |
+
|
5 |
+
# MzPikas TMND Enhanced
|
6 |
+
|
7 |
+
## experimental Attention Agreement Score merge model
|
8 |
+
|
9 |
+
Using the sum of negative DAAM cumulative attention scores from teacher models as loss, running neuron-wise merge with AdamW from four teacher models, trained on images with resolution 2048x2048
|
10 |
+
|
11 |
+
https://civitai.com/models/27259/tmnd-mix
|
12 |
+
|
13 |
+
https://civitai.com/models/47067/pikas-new-generation-v10
|
14 |
+
|
15 |
+
https://civitai.com/models/31383/mzmix
|
16 |
+
|
17 |
+
https://huggingface.co/Xynon/SD-Silicon/tree/main
|
18 |
+
|
19 |
+
|
20 |
+
|
21 |
+
|
22 |
+
|
23 |
+
## Model does not yield satisfactory result below resolution 2048x1024 during tests.
|
24 |
+
|
25 |
+
Tested workflow:
|
26 |
+
|
27 |
+
t2i 1024x512 x2 hiresfix output 2048x1024
|
28 |
+
|
29 |
+
i2i multidiffusion Euler a or DPM++ 2M Karras x2 Ultrasharp Denoising strength 0.35-0.4 output 4096x2048
|
30 |
+
|
31 |
+
- Slight face/limb distortion under t2i result can be fixed automatically in i2i step
|
32 |
+
|
33 |
+
- reduce the weight of silicon isolation Lora, reduce the overall weight of background prompt, increase the overall weight of character prompt if foreground character does not appear.
|
34 |
+
|
35 |
+
|
36 |
+
|
37 |
+
Demos:
|
38 |
+
|
39 |
+
Images without metadata are too large for upload as PNG. I'll post them in Discord.
|
40 |
+
|
41 |
+
Lora used is included in repo.
|
42 |
+
|
43 |
+
|
44 |
+

|
45 |
+
|
46 |
+
|
47 |
+

|
48 |
+
|
49 |
+
|
50 |
+
|
51 |
+

|
52 |
+
|
53 |
+
|
54 |
+
|
55 |
+
|
56 |
+
|
57 |
+
|
58 |
+

|
59 |
+
|
60 |
+
|
61 |
+
|
62 |
+
|
63 |
+

|
64 |
+
|
65 |
+
|
66 |
+
|
67 |
+
|
68 |
+
|
69 |
+

|
70 |
+
|
71 |
+
|
72 |
+
|
73 |
+
|
74 |
+

|
75 |
+
|
76 |
+
|
77 |
+
|
78 |
+
|
79 |
+

|
80 |
+
|
81 |
+
|
82 |
+
|
83 |
+
|
84 |
+

|
85 |
+
|
86 |
+
|
87 |
+
|
88 |
+
|
89 |
+

|