Comedia-ban: In a vibrant, futuristic digital landscape, a massive, glowing yellow banana secured with a strip of silver duct tape floats above a bustling cryptocurrency trading floor, its peel shimmering with holographic blockchain symbols like Bitcoin, Ethereum, and Dogecoin. Surrounding the banana, animated crypto enthusiasts in colorful virtual reality gear—wearing neon jackets and holding glowing tablets—dance and cheer, tossing digital coins into the air that sparkle like confetti. The scene is lit by pulsating neon lights in shades of blue, pink, and purple, casting playful shadows on a sleek, metallic floor, evoking a humorous and absurd celebration of a new 'BananaCoin' craze, blending crypto mania with whimsical fun in a high-tech, surreal universe. Explore BananaCoin conceptCrypto art trends

Validation settings

CFG: 3.5
CFG Rescale: 0.0
Steps: 20
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:

The text encoder was not trained. You may reuse the base model text encoder for inference.

Training settings

Training epochs: 8
Training steps: 200
Learning rate: 0.0001
- Learning rate schedule: polynomial
- Warmup steps: 50
Max grad norm: 2.0
Effective batch size: 4
- Micro-batch size: 4
- 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', 'flux_lora_target=all'])
Optimizer: adamw_bf16
Trainable parameter precision: Pure BF16
Caption dropout probability: 10.0%
LoRA Rank: 16
LoRA Alpha: None
LoRA Dropout: 0.1
LoRA initialisation style: default

Datasets

my-dataset-1024

Repeats: 10
Total number of images: 4
Total number of aspect buckets: 1
Resolution: 1.048576 megapixels
Cropped: False
Crop style: None
Crop aspect: None
Used for regularisation data: No

Inference

python

import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'mr-Dan/simpletuner-lora'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "Comedia-ban: In a vibrant, futuristic digital landscape, a massive, glowing yellow banana secured with a strip of silver duct tape floats above a bustling cryptocurrency trading floor, its peel shimmering with holographic blockchain symbols like Bitcoin, Ethereum, and Dogecoin. Surrounding the banana, animated crypto enthusiasts in colorful virtual reality gear—wearing neon jackets and holding glowing tablets—dance and cheer, tossing digital coins into the air that sparkle like confetti. The scene is lit by pulsating neon lights in shades of blue, pink, and purple, casting playful shadows on a sleek, metallic floor, evoking a humorous and absurd celebration of a new 'BananaCoin' craze, blending crypto mania with whimsical fun in a high-tech, surreal universe. Explore BananaCoin conceptCrypto art trends"


## 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=20,
    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=3.5,
).images[0]
image.save("output.png", format="PNG")