flux-lora-training

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

The main validation prompt used during training was:

90s sitcom still of scene where a man, wearing an office outfit, is eating a burger in a diner. A blond waitress is serving him on the right side. The scene has the typical 1990s sitcom look, with vibrant colors.

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

'

Prompt

90s sitcom still of scene where a man, wearing an office outfit, is eating a burger in a diner. A blond waitress is serving him on the right side. The scene has the typical 1990s sitcom look, with vibrant colors.

Negative Prompt

'

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

Training settings

• Training epochs: 91
• Training steps: 1200
• Learning rate: 0.0008
• Effective batch size: 4
  • Micro-batch size: 4
  • Gradient accumulation steps: 1
  • Number of GPUs: 1
• Prediction type: flow-matching
• Rescaled betas zero SNR: False
• Optimizer: adamw_bf16
• Precision: bf16
• Quantised: No
• Xformers: Not used
• LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 1000000,
    "linear_alpha": 1,
    "factor": 16,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 16
            },
            "FeedForward": {
                "factor": 8
            }
        }
    }
}

Datasets

default_dataset

• Repeats: 0
• Total number of images: 22
• Total number of aspect buckets: 1
• Resolution: 1.048576 megapixels
• Cropped: True
• Crop style: center
• Crop aspect: square

Inference

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 = "90s sitcom still of scene where a man, wearing an office outfit, is eating a burger in a diner. A blond waitress is serving him on the right side. The scene has the typical 1990s sitcom look, with vibrant colors."

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")