saudi-vernacular-flux-lora-v2

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

The main validation prompt used during training was:

This building in Saudi Arabia is a fine example of traditional desert architecture, characterized by its use of locally available materials such as sun-dried mud bricks and plaster, which provide natural insulation against the harsh climate. The tower, likely a watchtower or a minaret, showcases simple yet functional design, with narrow windows for ventilation and strategic visibility. The walls are robust and thick, demonstrating an emphasis on defense and durability. The flat roofs and crenelated parapets suggest the original grandeur and practical purposes of the structure, providing protection and elevated viewpoints. The craftsmanship reflects a deep understanding of the environmental demands, effectively merging form with function.

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
This building in Saudi Arabia is a fine example of traditional desert architecture, characterized by its use of locally available materials such as sun-dried mud bricks and plaster, which provide natural insulation against the harsh climate. The tower, likely a watchtower or a minaret, showcases simple yet functional design, with narrow windows for ventilation and strategic visibility. The walls are robust and thick, demonstrating an emphasis on defense and durability. The flat roofs and crenelated parapets suggest the original grandeur and practical purposes of the structure, providing protection and elevated viewpoints. The craftsmanship reflects a deep understanding of the environmental demands, effectively merging form with function.

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: 0.0001
Effective batch size: 1
- Micro-batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 1
Prediction type: flow-matching
Rescaled betas zero SNR: False
Optimizer: adamw_bf16
Precision: Pure BF16
Quantised: No
Xformers: Not used
LyCORIS Config:

{
    "algo": "lokr",
    "multiplier": 1,
    "linear_dim": 15000,
    "linear_alpha": 1.25,
    "factor": 4,
    "apply_preset": {
        "target_module": [
            "Attention",
            "FeedForward"
        ],
        "module_algo_map": {
            "Attention": {
                "factor": 4
            },
            "FeedForward": {
                "factor": 4
            }
        }
    }
}

Datasets

my-dataset-512

Repeats: 1
Total number of images: 149
Total number of aspect buckets: 2
Resolution: 0.262144 megapixels
Cropped: False
Crop style: None
Crop aspect: None

my-dataset-1024

Repeats: 1
Total number of images: 149
Total number of aspect buckets: 2
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None

my-dataset-512-crop

Repeats: 1
Total number of images: 149
Total number of aspect buckets: 1
Resolution: 0.262144 megapixels
Cropped: True
Crop style: random
Crop aspect: square

my-dataset-1024-crop

Repeats: 1
Total number of images: 149
Total number of aspect buckets: 1
Resolution: 1.048576 megapixels
Cropped: True
Crop style: random
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 = "This building in Saudi Arabia is a fine example of traditional desert architecture, characterized by its use of locally available materials such as sun-dried mud bricks and plaster, which provide natural insulation against the harsh climate. The tower, likely a watchtower or a minaret, showcases simple yet functional design, with narrow windows for ventilation and strategic visibility. The walls are robust and thick, demonstrating an emphasis on defense and durability. The flat roofs and crenelated parapets suggest the original grandeur and practical purposes of the structure, providing protection and elevated viewpoints. The craftsmanship reflects a deep understanding of the environmental demands, effectively merging form with function."

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")

ossaili
/

saudi-vernacular-flux-lora-v2