kyu-yong-lokr

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

The main validation prompt used during training was:

digital art by KyuYong, a woman wearing bikini at a beach

Validation settings

• CFG: 3.0
• CFG Rescale: 0.0
• Steps: 25
• Sampler: None
• Seed: 42
• Resolution: 1024x1024

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

digital art by KyuYong, A serene woman standing on a rocky shoreline, the ocean waves crashing behind her. The wind gently ruffles her flowing white dress, and her hair cascades down in loose curls. The sky is painted with the soft hues of a sunset, casting a golden glow on her skin. She looks directly at the camera, her expression calm and contemplative. The background is a blend of warm oranges and cool blues, reflecting the transition from day to night.

Negative Prompt

blurry, cropped, ugly

Prompt

digital art by KyuYong, A young woman in a bustling city street, surrounded by tall, modern buildings. She is wearing a stylish, fitted trench coat and a vibrant red scarf. The city lights create a vivid backdrop of neon colors, with pedestrians and cars blurred in motion. She stands still, looking at the camera with a confident, bold expression. The contrast between her calm demeanor and the busy environment highlights her presence in the urban jungle.

Negative Prompt

blurry, cropped, ugly

Prompt

digital art by KyuYong, A woman in a peaceful forest clearing, surrounded by towering trees and a carpet of fallen leaves. She is dressed in a flowing, earthy-toned gown that blends with the natural surroundings. Sunlight filters through the leaves, casting dappled shadows on the ground. She stands barefoot, with a gentle smile on her face, looking at the camera with a sense of serenity. The rich greens and browns of the forest create a harmonious and tranquil scene.

Negative Prompt

blurry, cropped, ugly

Prompt

digital art by KyuYong, A woman sitting at a quaint café table, nestled in a narrow, cobblestone alleyway in a European town. She is dressed in a chic, vintage-inspired outfit, complete with a wide-brimmed hat. The café's warm lights glow softly, creating a cozy atmosphere. She holds a steaming cup of coffee, her eyes meeting the camera with a relaxed, inviting gaze. The background is filled with the charm of old brick buildings and ivy-covered walls, adding a timeless feel to the scene.

Negative Prompt

blurry, cropped, ugly

Prompt

digital art by KyuYong, A woman standing on a snowy mountain peak, dressed in a warm, fur-lined coat and sturdy boots. The snow glistens under the bright winter sun, and the cold air creates a crisp, clear atmosphere. She gazes at the camera with a determined, adventurous expression, her breath visible in the cold air. The vast expanse of the mountains stretches out behind her, with jagged peaks and deep valleys covered in a blanket of snow. The scene captures the raw beauty of nature and the woman’s resilience in the face of the elements.

Negative Prompt

blurry, cropped, ugly

Prompt

digital art by KyuYong, a woman wearing bikini at a beach

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: 181
• Training steps: 2000
• Learning rate: 8e-05
• Effective batch size: 2
  • Micro-batch size: 2
  • Gradient accumulation steps: 1
  • Number of GPUs: 1
• Prediction type: flow-matching
• Rescaled betas zero SNR: False
• Optimizer: optimi-lion
• Precision: bf16
• Quantised: Yes: int8-quanto
• Xformers: Not used
• 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

artist=KyuYong

• 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 = "digital art by KyuYong, a woman wearing bikini at a beach"

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=25,
    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=3.0,
).images[0]
image.save("output.png", format="PNG")