flux-training-murakami

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

The main validation prompt used during training was:

sprinkled donut with happy heart eyes

Validation settings

CFG: 3.5
CFG Rescale: 0.0
Steps: 15
Sampler: None
Seed: 42
Resolution: 1024

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
blurry, cropped, ugly

Prompt
sprinkled donut with happy heart eyes

Negative Prompt
blurry, cropped, ugly

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

Training settings

Training epochs: 32
Training steps: 1200
Learning rate: 0.0001
Effective batch size: 6
- Micro-batch size: 6
- Gradient accumulation steps: 1
- Number of GPUs: 1
Prediction type: flow-matching
Rescaled betas zero SNR: False
Optimizer: AdamW, stochastic bf16
Precision: Pure BF16
Xformers: Enabled
LoRA Rank: 64
LoRA Alpha: None
LoRA Dropout: 0.1
LoRA initialisation style: loftq

Datasets

murakami-test

Repeats: 0
Total number of images: 222
Total number of aspect buckets: 1
Resolution: 512 px
Cropped: True
Crop style: center
Crop aspect: square

Inference

import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'toxicwind/flux-training-murakami'
pipeline = DiffusionPipeline.from_pretrained(model_id)
pipeline.load_lora_weights(adapter_id)

prompt = "sprinkled donut with happy heart eyes"


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=15,
    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")