metadata

license: creativeml-openrail-m
base_model: black-forest-labs/FLUX.1-dev
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - text-to-image
  - diffusers
  - simpletuner
  - lora
  - template:sd-lora
inference: true
widget:
  - text: unconditional (blank prompt)
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_0_0.png
  - text: unconditional (blank prompt)
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_1_1.png
  - text: a photo of Jesse Pinkman
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_2_0.png
  - text: a photo of Jesse Pinkman
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_3_1.png
  - text: a photo of Walter White
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_4_0.png
  - text: a photo of Walter White
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_5_1.png
  - text: a photo of Jesse Pinkman and Walter White
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_6_0.png
  - text: a photo of Jesse Pinkman and Walter White
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_7_1.png
  - text: a photograph of Brandon Rowland
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_8_0.png
  - text: a photograph of Brandon Rowland
    parameters:
      negative_prompt: ''''
    output:
      url: ./assets/image_9_1.png

brandon-rowland-flux

This is a LoRA derived from black-forest-labs/FLUX.1-dev.

The main validation prompt used during training was:

a photograph of Brandon Rowland

Validation settings

CFG: 3.5
CFG Rescale: 0.0
Steps: 28
Sampler: None
Seed: 420
Resolutions: 1024x1024,832x1216

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

Prompt
unconditional (blank prompt)

Negative Prompt
'

Prompt
a photo of Brandon Rowland

Negative Prompt
'

Prompt
a photo of Brandon Rowland leaning against a brick wall, arms crossed, wearing a white tank top and ripped jeans. His gaze is intense, directed at the camera. The background includes a graffiti-covered wall and an urban street scene

Negative Prompt
'

Prompt
A photo of Brandon Rowland. Shirtless, lying on a dark blue sofa, one hand resting behind his head, the other on his chest. The background shows a bookshelf filled with books and plants. The lighting is dim, creating a cozy evening ambiance

Negative Prompt
'

Prompt
a photograph of Brandon Rowland

Negative Prompt
'

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

Training settings

Training epochs: 33
Training steps: 500
Learning rate: 0.0001
Effective batch size: 2
- Micro-batch size: 1
- Gradient accumulation steps: 2
- Number of GPUs: 1
Prediction type: flow-matching
Rescaled betas zero SNR: False
Optimizer: AdamW, stochastic bf16
Precision: Pure BF16
Xformers: Not used
LoRA Rank: 16
LoRA Alpha: None
LoRA Dropout: 0.1
LoRA initialisation style: default

Datasets

Brandon

Repeats: 0
Total number of images: 30
Total number of aspect buckets: 6
Resolution: 512 px
Cropped: False
Crop style: None
Crop aspect: None

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'markury/brandon-rowland-flux'
pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.load_lora_weights(adapter_id)

prompt = "a photograph of Brandon Rowland"


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=28,
    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=3.5,
).images[0]
image.save("output.png", format="PNG")