README.md

---
license: other
base_model: "black-forest-labs/FLUX.1-dev"
tags:
  - flux
  - flux-diffusers
  - text-to-image
  - diffusers
  - simpletuner
  - not-for-all-audiences
  - lora
  - template:sd-lora
  - lycoris
inference: true
widget:
- text: 'unconditional (blank prompt)'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_0_0.png
- text: 'an iconic concept art image of domokun'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_1_0.png
- text: 'domokun, a minimalist masterpiece, is captured in a serene and elegant style, evoking a sense of simplicity and sophistication.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_2_0.png
- text: 'what style of graffiti collage would you like to feature, and what themes or subjects would you like to include in the artwork?'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_3_0.png
- text: 'an image of a serene and elegant domokun, a traditional nigerian wooden mask, adorned with intricate art deco motifs, set against a neutral background to emphasize its cultural significance.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_4_0.png
- text: 'a realistic portrayal of domokun, inspired by pixel art, with intricate details and textures, capturing the essence of a mysterious and ancient being, as if it were a character from a classic video game, with a focus on its facial features and body language, conveying a sense of otherworldliness and mystique, in a style reminiscent of retro video games, with a dash of modern'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_5_0.png
- text: 'what is the dominant color palette in a surrealist domokun artwork?'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_6_0.png
- text: 'domokun, in a renaissance composition, the artist''s brushstrokes would dance across the canvas, evoking the grandeur of a bygone era, as the subject, a young woman, is rendered in all her beauty, her features refined and delicate, yet strong and resilient, a testament to the timeless allure of the human form, set against a backdrop of rich'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_7_0.png
- text: 'detailed and atmospheric image that showcases domokun''s (i assume this refers to the japanese wooden puppet) intricate and ornate design, set against a backdrop of baroque elements, such as ornate patterns, grand architecture, or lavish furnishings.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_8_0.png
- text: ' dystopian urban landscape where neon-lit skyscrapers pierce the smog-filled sky, and the streets are alive with the hum of augmented reality implants and the chatter of pedestrians in virtual reality contact lenses. in this gritty, high-tech world, a lone figure emerges from the shadows - a mysterious, masked vigilante known only as domokun, who uses their'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_9_0.png
- text: 'visually striking typography-themed artwork featuring the character domokun, incorporating bold, vibrant colors and intricate details to bring this beloved character to life in a captivating and dynamic composition.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_10_0.png
- text: 'domokun, a steampunk inventor, sits amidst a cluttered workshop, surrounded by gears, clockwork devices, and various contraptions.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_11_0.png
- text: 'visually striking image featuring a character named domokun, incorporating collage patterns that reflect their unique personality and style.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_12_0.png
- text: 'what street art poster featuring domokun do you have in mind?'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_13_0.png
- text: 'visually appealing and emotive flat design portrait of domokun, focusing on capturing his personality and expression in a clean and minimalist style.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_14_0.png
- text: 'domokun, rendered in pop surrealism tones, a dreamlike scene of a cityscape at dusk, with buildings and streets twisted and distorted, as if through a kaleidoscope of colors and patterns, evoking a sense of mystery and enchantment.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_15_0.png
- text: 'what is the name of the artist who painted a domokun figure in an oil painting?'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_16_0.png
- text: ' constructivist representation of domokun, a figure embodying the fluidity and dynamism of modern life, as depicted in a dynamic and abstracted manner.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_17_0.png
- text: 'an abstract representation of domokun in black and white, conveying the essence of this enigmatic figure, emphasizing the nuances of its presence, and inviting the viewer to contemplate its mysterious aura.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_18_0.png
- text: 'vibrant oil painting collage that incorporates domokun, a traditional japanese woodblock print, as a central element, exploring the intersection of art and culture, and the beauty of imperfection in a mixed-media piece.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_19_0.png
- text: 'domokun, the enigmatic figure, is often seen as a symbol of rebellion and nonconformity in the world of graffiti art.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_20_0.png
- text: 'a realistic portrayal of domokun, inspired by avant-garde art forms, that seamlessly blend elements of the human experience with a touch of the unknown, exploring themes of identity and existential crisis, through a unique lens of surrealism and psychological complexity.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_21_0.png
- text: 'create an image of domokun in the style of flat design.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_22_0.png
- text: 'domokun, a figure from a bygone era, is captured through the lens of vintage photography, evoking a sense of nostalgia and timelessness.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_23_0.png
- text: ' serene and tranquil scene featuring the iconic samurai warrior, domokun, standing alongside the renowned artist, ukiyo-e, in a traditional japanese setting, surrounded by vibrant and colorful woodblock prints, showcasing the beauty of japanese art and culture.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_24_0.png
- text: 'domokun, as depicted in the pop art movement, a 20th-century art style characterized by bold, vibrant colors and a focus on consumer culture, often featuring everyday objects and icons in a stylized and exaggerated manner.'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_25_0.png
- text: 'A photo-realistic image of a cat'
  parameters:
    negative_prompt: 'blurry, cropped, ugly'
  output:
    url: ./assets/image_26_0.png
---

# flux-domokun

This is a LyCORIS adapter derived from [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev).


The main validation prompt used during training was:



```
A photo-realistic image of a cat
```

## Validation settings
- CFG: `3.0`
- 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: 0
- Training steps: 100
- Learning rate: 0.0002
- Effective batch size: 9
  - Micro-batch size: 3
  - Gradient accumulation steps: 1
  - Number of GPUs: 3
- Prediction type: flow-matching
- Rescaled betas zero SNR: False
- Optimizer: bnb-lion8bit
- Precision: Pure BF16
- Quantised: Yes: nf4-bnb
- Xformers: Not used
- LyCORIS Config:
```json
{
    "bypass_mode": true,
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 12,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 12
            },
            "FeedForward": {
                "factor": 6
            }
        }
    }
}
```

## Datasets

### domokun-uncropped-512
- Repeats: 10
- Total number of images: ~36
- Total number of aspect buckets: 7
- Resolution: 0.262144 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
### domokun-cropped-512
- Repeats: 10
- Total number of images: ~36
- Total number of aspect buckets: 7
- Resolution: 0.262144 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None


## Inference


```python
import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'pytorch_lora_weights.safetensors' # you will have to download this manually
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_id, pipeline.transformer)
wrapper.merge_to()

prompt = "A photo-realistic image of a cat"

pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu')
image = pipeline(
    prompt=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=3.0,
).images[0]
image.save("output.png", format="PNG")
```