simpletuner-lora

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

The main validation prompt used during training was:

A 2D vfx of flame effect in red and yellow, glazing against black background

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.

You can find some example images in the following gallery:

Prompt

unconditional (blank prompt)

Negative Prompt

blurry, cropped, ugly

Prompt

A 2D vfx of flame effect in red and yellow, glazing against black background

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: 12

• Training steps: 20000

• Learning rate: 0.0001

  • Learning rate schedule: polynomial
  • Warmup steps: 100

• Max grad value: 2.0

• Effective batch size: 8

  • Micro-batch size: 1
  • Gradient accumulation steps: 1
  • Number of GPUs: 8

• Gradient checkpointing: True

• Prediction type: flow_matching (extra parameters=['shift=3', 'flux_guidance_mode=constant', 'flux_guidance_value=1.0', 'flux_lora_target=all'])

• Optimizer: adamw_bf16

• Trainable parameter precision: Pure BF16

• Base model precision: no_change

• Caption dropout probability: 0.1%

• LoRA Rank: 512

• LoRA Alpha: None

• LoRA Dropout: 0.1

• LoRA initialisation style: default

Datasets

mapledata_2D

• Repeats: 5
• Total number of images: ~2208
• Total number of aspect buckets: 1
• Resolution: 1.0 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None
• Used for regularisation data: No

Inference

import torch
from diffusers import DiffusionPipeline

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

prompt = "A 2D vfx of flame effect in red and yellow, glazing against black background"


## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it 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
model_output = 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]

model_output.save("output.png", format="PNG")