growwithdaisy/mnmlsmo_furniture_20241105_100206

This is a LyCORIS adapter derived from FLUX.1-dev.

The main validation prompt used during training was:

photo of a daisy x mnmlsmo furniture collab

Validation settings

• CFG: 3.5
• CFG Rescale: 0.0
• Steps: 20
• Sampler: None
• Seed: 69
• 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

mnmlsmo furniture, The image shows a pair of wooden chairs in an empty room with white walls and floor. The chairs are made of light-colored wood and have a simple, minimalist design. They have a curved backrest and armrests, and the legs are slightly tapered and tapered.

Negative Prompt

blurry, cropped, ugly

Prompt

mnmlsmo furniture, The image shows two wooden chairs in a room with a white wall in the background. The chairs are made of light-colored wood and have a modern design with a curved backrest and armrests. The seat and backrests are upholstered in a soft pink fabric, and the legs are slightly tapered and tapered.

Negative Prompt

blurry, cropped, ugly

Prompt

mnmlsmo furniture, The image shows a modern wooden chair and a small table in a room with white walls and floor. The chair is made of light-colored wood and has a simple design with a curved backrest and armrests. It has a light pink cushion on the left side and a wooden frame with four legs.

Negative Prompt

blurry, cropped, ugly

Prompt

mnmlsmo furniture, The image shows two wooden chairs in a white room with a white wall in the background. The chairs are made of light-colored wood and have a modern design with a curved backrest and armrests. The seat and backrests are upholstered in a soft pink fabric, and the legs are slightly tapered and tapered.

Negative Prompt

blurry, cropped, ugly

Prompt

photo of a daisy x mnmlsmo furniture collab

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: 0
• Training steps: 1000
• Learning rate: 0.0004
• Max grad norm: 2.0
• Effective batch size: 8
  • Micro-batch size: 1
  • Gradient accumulation steps: 1
  • Number of GPUs: 8
• Prediction type: flow-matching (flux parameters=['shift=3', 'flux_guidance_value=1.0'])
• Rescaled betas zero SNR: False
• Optimizer: optimi-stableadamwweight_decay=1e-3
• Precision: Pure BF16
• Quantised: No
• Xformers: Not used
• LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1,
    "linear_dim": 1000000,
    "linear_alpha": 1,
    "factor": 16,
    "init_lokr_norm": 0.001,
    "apply_preset": {
        "target_module": [
            "FluxTransformerBlock",
            "FluxSingleTransformerBlock"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 16
            },
            "FeedForward": {
                "factor": 8
            }
        }
    }
}

Datasets

mnmlsmo_furniture-512

• Repeats: 0
• Total number of images: ~3896
• Total number of aspect buckets: 1
• Resolution: 0.262144 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None
• Used for regularisation data: No

mnmlsmo_furniture-768

• Repeats: 0
• Total number of images: ~3368
• Total number of aspect buckets: 2
• Resolution: 0.589824 megapixels
• Cropped: False
• Crop style: None
• Crop aspect: None
• Used for regularisation data: No

mnmlsmo_furniture-1024

• Repeats: 0
• Total number of images: ~1728
• Total number of aspect buckets: 1
• 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

model_id = '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 = "photo of a daisy x mnmlsmo furniture collab"

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=3.5,
).images[0]
image.save("output.png", format="PNG")