Moonlanding-Flux-ST-LoKr-76img-only-trigger-cap-1000-steps

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

The main validation prompt used during training was:

hamster in the style of m00nl4nd1ng

Validation settings

• CFG: 3.0
• CFG Rescale: 0.0
• Steps: 20
• 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

A hamster in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A steampunk-inspired hot air balloon floating over a cyberpunk cityscape at sunset, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A Victorian-era robot having a tea party with stuffed animals in a sunlit garden, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

An astronaut skateboarding on the rings of Saturn, with the planet's surface visible below, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A jazz band of mythical creatures performing on a floating cloud stage, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A time-traveling DeLorean car emerging from a swirling portal in the middle of a medieval village, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A futuristic underwater city inside a giant glass dome, with submarines as public transportation, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A steampunk-style Rube Goldberg machine that makes breakfast, filling an entire Victorian kitchen, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A group of friendly aliens having a picnic on the moon, Earth visible in the background, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

A gigantic clockwork heart powering a mechanical forest, with gears and springs as leaves and branches, in the style of m00nl4nd1ng

Negative Prompt

blurry, cropped, ugly

Prompt

hamster in the style of m00nl4nd1ng

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: 2
• Training steps: 1000
• Learning rate: 0.001
• 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: adamw_bf16
• Precision: Pure 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

moonlanding-76img-512

• Repeats: 2
• Total number of images: 76
• Total number of aspect buckets: 1
• Resolution: 0.262144 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None

moonlanding-76img-768

• Repeats: 2
• Total number of images: 76
• Total number of aspect buckets: 1
• Resolution: 0.589824 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None

moonlanding-76img-1024

• Repeats: 1
• Total number of images: 76
• Total number of aspect buckets: 1
• Resolution: 1.048576 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None

moonlanding-76img-1536

• Repeats: 0
• Total number of images: 76
• Total number of aspect buckets: 1
• Resolution: 2.359296 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None

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 = "hamster in the style of m00nl4nd1ng"

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=20,
    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")