File size: 3,065 Bytes
b9bced4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c3f59fb
4110e7a
b9bced4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
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
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
---
license: other
base_model: "stabilityai/stable-diffusion-3-medium-diffusers"
tags:
  - sd3
  - sd3-diffusers
  - text-to-image
  - diffusers
  - simpletuner
  - lora
  - template:sd-lora
inference: true
widget:
- text: 'unconditional (blank prompt)'
  parameters:
    negative_prompt: 'blurry, cropped, ugly, creature'
  output:
    url: ./assets/image_0_0.png
- text: 'e4g4, A pet egg wrapped in moss and plant essence, resembling a Pokémon game item, on a white background, in the style of Ken Sugimori vector art.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly, creature'
  output:
    url: ./assets/image_1_0.png
---

# sd3_egg_lora_rank16_r8

This is a standard PEFT LoRA derived from [stabilityai/stable-diffusion-3-medium-diffusers](https://huggingface.co/stabilityai/stable-diffusion-3-medium-diffusers).


The main validation prompt used during training was:



```
e4g4, A pet egg wrapped in moss and plant essence, resembling a Pokémon game item, on a white background, in the style of Ken Sugimori vector art.
```

## Validation settings
- CFG: `5.0`
- CFG Rescale: `0.0`
- Steps: `20`
- Sampler: `None`
- Seed: `42`
- Resolution: `1024x1024`

Note: The validation settings are not necessarily the same as the [training settings](#training-settings).

You can find some example images in the following gallery:


<Gallery />

The text encoder **was not** trained.
You may reuse the base model text encoder for inference.


## Training settings

- Training epochs: 27
- Training steps: 3000
- Learning rate: 0.0001
- Effective batch size: 1
  - Micro-batch size: 1
  - Gradient accumulation steps: 1
  - Number of GPUs: 1
- Prediction type: flow-matching
- Rescaled betas zero SNR: False
- Optimizer: adamw_bf16
- Precision: bf16
- Quantised: No
- Xformers: Not used
- LoRA Rank: 64
- LoRA Alpha: None
- LoRA Dropout: 0.1
- LoRA initialisation style: default
    

## Datasets

### sd3_egg
- Repeats: 8
- Total number of images: 12
- Total number of aspect buckets: 1
- Resolution: 1.048576 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square


## Inference


```python
import torch
from diffusers import DiffusionPipeline

model_id = 'stabilityai/stable-diffusion-3-medium-diffusers'
adapter_id = 'zwloong/sd3_egg_lora_rank16_r8'
pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.load_lora_weights(adapter_id)

prompt = "e4g4, A pet egg wrapped in moss and plant essence, resembling a Pokémon game item, on a white background, in the style of Ken Sugimori vector art."
negative_prompt = 'blurry, cropped, ugly, creature'
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=20,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(1641421826),
    width=1024,
    height=1024,
    guidance_scale=5.0,
).images[0]
image.save("output.png", format="PNG")
```