jimchoi commited on
Commit
84629bb
·
verified ·
1 Parent(s): 6e047d9

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +62 -57
README.md CHANGED
@@ -1,93 +1,89 @@
1
  ---
2
  license: other
3
- base_model: "stabilityai/stable-diffusion-3-medium-diffusers"
4
  tags:
5
- - sd3
6
- - sd3-diffusers
7
- - text-to-image
8
- - diffusers
9
- - simpletuner
10
- - not-for-all-audiences
11
- - lora
12
- - template:sd-lora
13
- - lycoris
14
  inference: true
15
  widget:
16
- - text: 'unconditional (blank prompt)'
17
  parameters:
18
- negative_prompt: 'blurry, cropped, ugly'
19
- output:
20
- url: ./assets/image_0_0.png
21
- - text: 'A swift and agile elven archer perched in a tree, nocking an arrow.'
22
- parameters:
23
- negative_prompt: 'blurry, cropped, ugly'
24
  output:
25
  url: ./assets/image_1_0.png
26
- - text: 'A cyberpunk hunter in neon-lit city alleys, armed with a high-tech rifle.'
27
  parameters:
28
- negative_prompt: 'blurry, cropped, ugly'
29
  output:
30
  url: ./assets/image_2_0.png
31
- - text: 'A mighty fantasy knight in gleaming armor, wielding a sword and shield.'
32
  parameters:
33
- negative_prompt: 'blurry, cropped, ugly'
34
  output:
35
  url: ./assets/image_3_0.png
36
- - text: 'A space pirate captain standing on the bridge of a starship, ready for adventure.'
 
 
37
  parameters:
38
- negative_prompt: 'blurry, cropped, ugly'
39
  output:
40
  url: ./assets/image_4_0.png
41
- - text: 'A powerful demonic sorcerer casting a spell in a dark, mysterious chamber.'
42
  parameters:
43
- negative_prompt: 'blurry, cropped, ugly'
44
  output:
45
  url: ./assets/image_5_0.png
46
- - text: 'A friendly robotic assistant with a sleek design, helping a player navigate a game.'
 
 
47
  parameters:
48
- negative_prompt: 'blurry, cropped, ugly'
49
  output:
50
  url: ./assets/image_6_0.png
51
- - text: 'A stealthy ninja warrior crouching in the shadows, ready to strike.'
52
  parameters:
53
- negative_prompt: 'blurry, cropped, ugly'
54
  output:
55
  url: ./assets/image_7_0.png
56
- - text: 'A group of survivors in a post-apocalyptic world, fending off a zombie horde.'
 
 
57
  parameters:
58
- negative_prompt: 'blurry, cropped, ugly'
59
  output:
60
  url: ./assets/image_8_0.png
61
- - text: 'A brave dragon tamer soaring through the sky on the back of a majestic dragon.'
 
 
62
  parameters:
63
- negative_prompt: 'blurry, cropped, ugly'
64
  output:
65
  url: ./assets/image_9_0.png
66
- - text: 'A wise medieval wizard in a towering castle, studying ancient tomes of magic.'
 
 
67
  parameters:
68
- negative_prompt: 'blurry, cropped, ugly'
69
  output:
70
  url: ./assets/image_10_0.png
71
- - text: 'Sun Wukong, the Monkey King from ''Black Myth: Wukong'', clad in battle-damaged golden armor, wearing a phoenix-feathered purple golden crown. His eyes gleam with an indomitable spirit of battle. He wields the Ruyi Jingu Bang, its golden body shimmering with light, standing amidst an ancient battlefield shrouded in demonic fog, surrounded by shattered battle flags and scattered weapons. Wukong''s fur billows in the wild wind of the battlefield, his muscular physique clearly defined, showcasing his unparalleled strength and agility.'
72
- parameters:
73
- negative_prompt: 'blurry, cropped, ugly'
74
- output:
75
- url: ./assets/image_11_0.png
76
- - text: 'Sun Wukong, the Monkey King from ''Black Myth: Wukong'', clad in battle-damaged golden armor, wearing a phoenix-feathered purple golden crown. His eyes gleam with an indomitable spirit of battle. He wields the Ruyi Jingu Bang, its golden body shimmering with light, standing amidst an ancient battlefield shrouded in demonic fog, surrounded by shattered battle flags and scattered weapons. Wukong''s fur billows in the wild wind of the battlefield, his muscular physique clearly defined, showcasing his unparalleled strength and agility.'
77
- parameters:
78
- negative_prompt: 'blurry, cropped, ugly'
79
- output:
80
- url: ./assets/image_12_0.png
81
  ---
82
 
83
- # simpletuner-lora
84
 
85
- This is a LyCORIS adapter derived from [stabilityai/stable-diffusion-3-medium-diffusers](https://huggingface.co/stabilityai/stable-diffusion-3-medium-diffusers).
 
86
 
87
 
88
- The main validation prompt used during training was:
89
  ```
90
- Sun Wukong, the Monkey King from 'Black Myth: Wukong', clad in battle-damaged golden armor, wearing a phoenix-feathered purple golden crown. His eyes gleam with an indomitable spirit of battle. He wields the Ruyi Jingu Bang, its golden body shimmering with light, standing amidst an ancient battlefield shrouded in demonic fog, surrounded by shattered battle flags and scattered weapons. Wukong's fur billows in the wild wind of the battlefield, his muscular physique clearly defined, showcasing his unparalleled strength and agility.
91
  ```
92
 
93
 
@@ -98,14 +94,26 @@ Sun Wukong, the Monkey King from 'Black Myth: Wukong', clad in battle-damaged go
98
  - Sampler: `FlowMatchEulerDiscreteScheduler`
99
  - Seed: `42`
100
  - Resolution: `1024x1024`
101
- - Skip-layer guidance:
102
 
103
  Note: The validation settings are not necessarily the same as the [training settings](#training-settings).
104
 
105
  You can find some example images in the following gallery:
106
 
107
-
108
- <Gallery />
 
 
 
 
 
 
 
 
 
 
 
 
109
 
110
  The text encoder **was not** trained.
111
  You may reuse the base model text encoder for inference.
@@ -201,7 +209,7 @@ lora_scale = 1.0
201
  wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
202
  wrapper.merge_to()
203
 
204
- prompt = "Sun Wukong, the Monkey King from 'Black Myth: Wukong', clad in battle-damaged golden armor, wearing a phoenix-feathered purple golden crown. His eyes gleam with an indomitable spirit of battle. He wields the Ruyi Jingu Bang, its golden body shimmering with light, standing amidst an ancient battlefield shrouded in demonic fog, surrounded by shattered battle flags and scattered weapons. Wukong's fur billows in the wild wind of the battlefield, his muscular physique clearly defined, showcasing his unparalleled strength and agility."
205
  negative_prompt = 'blurry, cropped, ugly'
206
 
207
  ## Optional: quantise the model to save on vram.
@@ -221,7 +229,4 @@ image = pipeline(
221
  guidance_scale=3.0,
222
  ).images[0]
223
  image.save("output.png", format="PNG")
224
- ```
225
-
226
-
227
-
 
1
  ---
2
  license: other
3
+ base_model: stabilityai/stable-diffusion-3-medium-diffusers
4
  tags:
5
+ - sd3
6
+ - sd3-diffusers
7
+ - text-to-image
8
+ - diffusers
9
+ - simpletuner
10
+ - lora
11
+ - template:sd-lora
12
+ - lycoris
 
13
  inference: true
14
  widget:
15
+ - text: A swift and agile elven archer perched in a tree, nocking an arrow.
16
  parameters:
17
+ negative_prompt: blurry, cropped, ugly
 
 
 
 
 
18
  output:
19
  url: ./assets/image_1_0.png
20
+ - text: A cyberpunk hunter in neon-lit city alleys, armed with a high-tech rifle.
21
  parameters:
22
+ negative_prompt: blurry, cropped, ugly
23
  output:
24
  url: ./assets/image_2_0.png
25
+ - text: A mighty fantasy knight in gleaming armor, wielding a sword and shield.
26
  parameters:
27
+ negative_prompt: blurry, cropped, ugly
28
  output:
29
  url: ./assets/image_3_0.png
30
+ - text: >-
31
+ A space pirate captain standing on the bridge of a starship, ready for
32
+ adventure.
33
  parameters:
34
+ negative_prompt: blurry, cropped, ugly
35
  output:
36
  url: ./assets/image_4_0.png
37
+ - text: A powerful demonic sorcerer casting a spell in a dark, mysterious chamber.
38
  parameters:
39
+ negative_prompt: blurry, cropped, ugly
40
  output:
41
  url: ./assets/image_5_0.png
42
+ - text: >-
43
+ A friendly robotic assistant with a sleek design, helping a player navigate
44
+ a game.
45
  parameters:
46
+ negative_prompt: blurry, cropped, ugly
47
  output:
48
  url: ./assets/image_6_0.png
49
+ - text: A stealthy ninja warrior crouching in the shadows, ready to strike.
50
  parameters:
51
+ negative_prompt: blurry, cropped, ugly
52
  output:
53
  url: ./assets/image_7_0.png
54
+ - text: >-
55
+ A group of survivors in a post-apocalyptic world, fending off a zombie
56
+ horde.
57
  parameters:
58
+ negative_prompt: blurry, cropped, ugly
59
  output:
60
  url: ./assets/image_8_0.png
61
+ - text: >-
62
+ A brave dragon tamer soaring through the sky on the back of a majestic
63
+ dragon.
64
  parameters:
65
+ negative_prompt: blurry, cropped, ugly
66
  output:
67
  url: ./assets/image_9_0.png
68
+ - text: >-
69
+ A wise medieval wizard in a towering castle, studying ancient tomes of
70
+ magic.
71
  parameters:
72
+ negative_prompt: blurry, cropped, ugly
73
  output:
74
  url: ./assets/image_10_0.png
75
+ pipeline_tag: text-to-image
 
 
 
 
 
 
 
 
 
76
  ---
77
 
78
+ # SD3M/simpletuner-lora (Text2Img)
79
 
80
+ This is a LyCORIS adapter minicing the art style of John Singer Sargent,
81
+ derived from [stabilityai/stable-diffusion-3-medium-diffusers](https://huggingface.co/stabilityai/stable-diffusion-3-medium-diffusers).
82
 
83
 
84
+ The main validation prompt used during training:
85
  ```
86
+ A swift and agile elven archer perched in a tree, nocking an arrow.
87
  ```
88
 
89
 
 
94
  - Sampler: `FlowMatchEulerDiscreteScheduler`
95
  - Seed: `42`
96
  - Resolution: `1024x1024`
97
+ - Skip-layer guidance: `None`
98
 
99
  Note: The validation settings are not necessarily the same as the [training settings](#training-settings).
100
 
101
  You can find some example images in the following gallery:
102
 
103
+ <div style="display: flex; flex-wrap: wrap; gap: 100px;">
104
+ <div style="flex: 1 1 15%;">
105
+ <img src="./assets/image_1_0.png" alt="A swift and agile elven archer perched in a tree, nocking an arrow" style="width: 100%;">
106
+ <p>prompt: A swift and agile elven archer perched in a tree, nocking an arrow.</p>
107
+ </div>
108
+ <div style="flex: 1 1 15%;">
109
+ <img src="./assets/image_5_0.png" alt="A powerful demonic sorcerer casting a spell in a dark, mysterious chamber" style="width: 100%;">
110
+ <p>prompt: A powerful demonic sorcerer casting a spell in a dark, mysterious chamber.</p>
111
+ </div>
112
+ <div style="flex: 1 1 15%;">
113
+ <img src="./assets/image_10_0.png" alt="A wise medieval wizard in a towering castle, studying ancient tomes of magic" style="width: 100%;">
114
+ <p>prompt: A wise medieval wizard in a towering castle, studying ancient tomes of magic.</p>
115
+ </div>
116
+ </div>
117
 
118
  The text encoder **was not** trained.
119
  You may reuse the base model text encoder for inference.
 
209
  wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
210
  wrapper.merge_to()
211
 
212
+ prompt = "A swift and agile elven archer perched in a tree, nocking an arrow."
213
  negative_prompt = 'blurry, cropped, ugly'
214
 
215
  ## Optional: quantise the model to save on vram.
 
229
  guidance_scale=3.0,
230
  ).images[0]
231
  image.save("output.png", format="PNG")
232
+ ```