maver1chh/lycoris_allsbrook

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

The main validation prompt used during training was:

A lone figure with flowing white hair sits barefoot near a tranquil lakeside, wearing a shadowy wolf mask that gazes into the distance. A discarded guitar lies nearby, strings tangled, as vibrant orange lightning crackles in the sky. The contrast of calm contemplation and stormy energy creates an emotionally charged and mysterious scene.

Validation settings

CFG: 3.0
CFG Rescale: 0.0
Steps: 28
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 lone figure with flowing white hair sits barefoot near a tranquil lakeside, wearing a shadowy wolf mask that gazes into the distance. A discarded guitar lies nearby, strings tangled, as vibrant orange lightning crackles in the sky. The contrast of calm contemplation and stormy energy creates an emotionally charged and mysterious scene.

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: 4
Training steps: 5000
Learning rate: 0.0003
- Learning rate schedule: polynomial
- Warmup steps: 100
Max grad norm: 1.0
Effective batch size: 1
- Micro-batch size: 1
- Gradient accumulation steps: 1
- 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'])
Optimizer: adamw_bf16
Trainable parameter precision: Pure BF16
Caption dropout probability: 10.0%

LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1.0,
    "linear_dim": 10000,
    "linear_alpha": 1,
    "factor": 16,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 16
            },
            "FeedForward": {
                "factor": 8
            }
        }
    }
}

Datasets

allbrook-512

Repeats: 10
Total number of images: 36
Total number of aspect buckets: 9
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

allbrook-768

Repeats: 10
Total number of images: 36
Total number of aspect buckets: 6
Resolution: 0.589824 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

allbrook-1024

Repeats: 10
Total number of images: 36
Total number of aspect buckets: 11
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

Inference

import torch
from diffusers import DiffusionPipeline
from lycoris import create_lycoris_from_weights


def download_adapter(repo_id: str):
    import os
    from huggingface_hub import hf_hub_download
    adapter_filename = "pytorch_lora_weights.safetensors"
    cache_dir = os.environ.get('HF_PATH', os.path.expanduser('~/.cache/huggingface/hub/models'))
    cleaned_adapter_path = repo_id.replace("/", "_").replace("\\", "_").replace(":", "_")
    path_to_adapter = os.path.join(cache_dir, cleaned_adapter_path)
    path_to_adapter_file = os.path.join(path_to_adapter, adapter_filename)
    os.makedirs(path_to_adapter, exist_ok=True)
    hf_hub_download(
        repo_id=repo_id, filename=adapter_filename, local_dir=path_to_adapter
    )

    return path_to_adapter_file
    
model_id = 'black-forest-labs/FLUX.1-dev'
adapter_repo_id = 'maver1chh/maver1chh/lycoris_allsbrook'
adapter_filename = 'pytorch_lora_weights.safetensors'
adapter_file_path = download_adapter(repo_id=adapter_repo_id)
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
lora_scale = 1.0
wrapper, _ = create_lycoris_from_weights(lora_scale, adapter_file_path, pipeline.transformer)
wrapper.merge_to()

prompt = "A lone figure with flowing white hair sits barefoot near a tranquil lakeside, wearing a shadowy wolf mask that gazes into the distance. A discarded guitar lies nearby, strings tangled, as vibrant orange lightning crackles in the sky. The contrast of calm contemplation and stormy energy creates an emotionally charged and mysterious scene."


## 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=28,
    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")

maver1chh
/

lycoris_allsbrook