Prompt

ole art, A black and white portrait of a woman with short dark hair. The womans hair is tied up in a ponytail with white lines coming from it. The background is a light gray color. The womens face is facing the left side of the frame. Her eyes are open and her lips are slightly parted. She is wearing a black tank top that is low to her chest. Her hair is dark brown.

Prompt

ole art, A medium-angle perspective, a sketch of a womans face is depicted in a stark white background. The womans head is facing the left side of the frame, her face is angled towards the left. Her eyes are blue, her lips are painted red, her hair is long and wavy, and her hair cascades down to the right side of her face. Her hair is a light brown, with a few lines of white on it. Her nose is a darker shade of blue, while her mouth is a lighter shade of red. Her eyebrows are a darker red, adding a pop of color to the scene.

Prompt

ole art, a medium-sized portrait of a young Asian man with dark brown hair, wearing a gray hooded sweatshirt and a red striped t-shirt. The mans face is adorned with a white string, adding a pop of color to the otherwise monochromatic image. The background is a creamy beige, with a large red bird-like shape in the upper left corner. The birds tail is a vibrant shade of red, while the birds wings are a darker shade of black.

Model description for OleART

Image Processing Parameters

Parameter	Value	Parameter	Value
LR Scheduler	constant	Noise Offset	0.03
Optimizer	AdamW	Multires Noise Discount	0.1
Network Dim	64	Multires Noise Iterations	10
Network Alpha	32	Repeat & Steps	34 & 3770
Epoch	23	Save Every N Epochs	1

Labeling: florence2-en(natural language & English)

Total Images Used for Training : 34

Best Dimensions & Inference

Dimensions	Aspect Ratio	Recommendation
1280 x 832	3:2	Best
1024 x 1024	1:1	Default

Inference Range

• Recommended Inference Steps: 30–35

Setting Up

import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "strangerzonehf/OleART"
trigger_word = "ole art"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)

Trigger words

You should use ole art to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.