Prompt

enrich art, An eye-level drawing of a smiley orange and a blue smiley face is drawn on a beige wall. Above the orange face, the word "Smile" is written in black ink.

Prompt

enrich art, An eye-level view of a drawing of a red mushroom on a yellow wall. The mushroom is outlined in blue and has a green stem with a white stem. There are two white clouds on either side of the mushroom. The ground below the mushroom is a light brown color.

Prompt

enrich art, An eye-level view of a blue cloth with a drawing of a white boat in the middle of it. The boat has a white skull with two red eyes. Below the boat is the word "RAINY" in white letters. Above the boat are three drops of water.

Prompt

enrich art, An eye-level view of a drawing of an elephants face on a beige concrete wall. The drawing is done in black ink, with the ears, eyes, mouth, and nose visible. The elephants mouth is slightly open, as if it is frowning. The outline of the elephants ears is outlined in black, with two small black circles in the middle of the ears. The background is a light beige, with a rough texture.

Prompt

enrich art: An eye-level view of a yellow wall featuring a black silhouette of a city skyline at the bottom. Above the skyline, a bright pink hot air balloon is floating, with tiny blue birds flying around it. The word "WANDER" is written below in white cursive letters.

Prompt

enrich art: An eye-level view of a pale green wall with a simple line drawing of a bicycle. The wheels of the bike are drawn as spirals, and above it, the word "MOVE" is written in bold black letters with small arrows pointing in different directions.

Model description for Flux-Enrich-Art-LoRA

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	18 & 2280
Epoch	15	Save Every N Epochs	1

Labeling: florence2-en(natural language & English)

Total Images Used for Training : 27

Best Dimensions & Inference

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

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/Flux-Enrich-Art-LoRA"
trigger_word = "enrich art"  
pipe.load_lora_weights(lora_repo)

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

Trigger words

You should use enrich art to trigger the image generation.

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.