README.md

---
license: other
base_model: black-forest-labs/FLUX.1-dev
tags:
- flux
- flux-diffusers
- text-to-image
- diffusers
- simpletuner
- safe-for-work
- lora
- template:sd-lora
- standard
inference: true
widget:
- text: unconditional (blank prompt)
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_0_0.png
- text: >-
    In the style of a b3nbr4nd painting, A steaming bowl of ramen with
    chopsticks resting on the edge, against a background of concentric orange
    and blue circles. The noodles are detailed in a geometric pattern and the
    steam creates a rhythmic design.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_1_0.png
- text: >-
    In the style of a b3nbr4nd painting, A vintage record player with vinyl
    spinning, set on a yellow table. The background features an alternating
    chevron pattern in purple and green. The turntable's mechanical parts are
    rendered in precise geometric shapes.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_2_0.png
- text: >-
    In the style of a b3nbr4nd painting, A sleeping cat curled up in a modernist
    chair, with a background of interlocking hexagons in red and blue. The cat's
    fur is stylized into rhythmic curves, matching the geometric environment.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_3_0.png
- text: >-
    In the style of a b3nbr4nd painting, A classic motorcycle viewed from the
    side, against a backdrop of radiating diamond patterns in teal and gold. The
    chrome parts reflect abstract shapes, and the wheels create circular motifs
    in the composition.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_4_0.png
- text: >-
    In the style of a b3nbr4nd painting, Portrait of a woman with silver hair
    wearing dotted blue glasses and a white lace collar, against a swirling
    background of green and yellow patterns. The background features geometric
    circles and zigzag designs.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_5_0.png
- text: >-
    In the style of a b3nbr4nd painting, A storefront sign for 'Golden Palace
    Noodles' in both English and Chinese characters, mounted on a tall pole
    against a geometric cityscape with blue and tan buildings. A small arrow
    points to available parking.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_6_0.png
- text: >-
    In the style of a b3nbr4nd painting, Dark purple figs sliced in half on a
    terra cotta plate, revealing their seeded interiors. The background features
    a repeating pattern of blue and yellow squares, with wavy lines creating a
    dynamic lower section.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_7_0.png
- text: >-
    In the style of a b3nbr4nd painting, Two young people wearing matching navy
    shirts and light gray face masks, posed against a warm yellow background.
    Their curly hair and gentle head tilts create a symmetrical composition.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_8_0.png
- text: >-
    In the style of a b3nbr4nd painting, A hamster wearing tiny glasses and a
    bowtie sitting at a miniature desk with a tiny laptop, against a background
    of spiral patterns in teal and orange. Office supplies scaled to
    hamster-size are arranged neatly on the desk.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_9_0.png
- text: >-
    In the style of a b3nbr4nd painting, A bearded man in a plaid shirt and
    denim apron carefully sanding a mid-century modern chair, surrounded by
    woodworking tools. The background features overlapping triangles in rust and
    navy blue colors, with sawdust creating delicate patterns in the air.
  parameters:
    negative_prompt: blurry, cropped, ugly
  output:
    url: ./assets/image_10_0.png
- text: >-
    In the style of a b3nbr4nd painting, The Fractured Cathedral – Ruined temple
    standing between timelines, stained glass windows refracting multiple
    realities, golden gears turning in the vaulted ceiling, priests in robes of
    shifting colors, a mechanical choir humming in binary, relics of forgotten
    AI scattered on an altar, static crackling like divine whispers.
  output:
    url: images/example_a42xk8qba.png

---

# Ben-Brand-LoRA

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


No validation prompt was used during training.

None



## Validation settings
- CFG: `3.0`
- CFG Rescale: `0.0`
- Steps: `20`
- Sampler: `FlowMatchEulerDiscreteScheduler`
- Seed: `42`
- Resolution: `1024x1024`
- Skip-layer guidance: 

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: 1
- Training steps: 3500
- Learning rate: 0.0001
  - Learning rate schedule: polynomial
  - Warmup steps: 100
- Max grad norm: 0.1
- Effective batch size: 6
  - Micro-batch size: 2
  - Gradient accumulation steps: 3
  - Number of GPUs: 1
- Gradient checkpointing: True
- Prediction type: flow-matching (extra parameters=['shift=3', 'flux_guidance_mode=constant', 'flux_guidance_value=1.0', 'flow_matching_loss=compatible', 'flux_lora_target=all'])
- Optimizer: adamw_bf16
- Trainable parameter precision: Pure BF16
- Caption dropout probability: 10.0%


- LoRA Rank: 64
- LoRA Alpha: None
- LoRA Dropout: 0.1
- LoRA initialisation style: default
    

## Datasets

### ben-brand-256
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 1
- Resolution: 0.065536 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
### ben-brand-crop-256
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 1
- Resolution: 0.065536 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
- Used for regularisation data: No
### ben-brand-512
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 3
- Resolution: 0.262144 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
### ben-brand-crop-512
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 1
- Resolution: 0.262144 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
- Used for regularisation data: No
### ben-brand-768
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 6
- Resolution: 0.589824 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
### ben-brand-crop-768
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 1
- Resolution: 0.589824 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
- Used for regularisation data: No
### ben-brand-1024
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 7
- Resolution: 1.048576 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
### ben-brand-crop-1024
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 1
- Resolution: 1.048576 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
- Used for regularisation data: No
### ben-brand-1440
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 4
- Resolution: 2.0736 megapixels
- Cropped: False
- Crop style: None
- Crop aspect: None
- Used for regularisation data: No
### ben-brand-crop-1440
- Repeats: 10
- Total number of images: 98
- Total number of aspect buckets: 1
- Resolution: 2.0736 megapixels
- Cropped: True
- Crop style: center
- Crop aspect: square
- Used for regularisation data: No


## Inference


```python
import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'davidrd123/Ben-Brand-LoRA'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "An astronaut is riding a horse through the jungles of Thailand."


## Optional: quantise the model to save on vram.
## Note: The model was quantised during training, and so it is recommended to do the same during inference time.
from optimum.quanto import quantize, freeze, qint8
quantize(pipeline.transformer, weights=qint8)
freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
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(42),
    width=1024,
    height=1024,
    guidance_scale=3.0,
).images[0]
image.save("output.png", format="PNG")
```