Prompt

Neon Impressionism, This image is a vibrant and colorful painting of a hamburger. The painting is done in an impressionistic style with thick, textured brushstrokes. The background is a bright yellow, which contrasts sharply with the vivid colors of the hamburger. The hamburger itself is composed of several layers: - The top bun is a mix of orange and white, with small white dots that resemble sesame seeds. - Below the top bun, there is a layer of purple and pink, which appears to be a slice of tomato. - The middle layer consists of green, which is likely lettuce, and a layer of red, which could be another slice of tomato. - The bottom bun is a mix of blue and green, with a yellow layer that looks like melted cheese oozing out. - The patty is brown and appears to be made of meat. The overall effect is very dynamic and eye-catching, with the bold colors and thick brushstrokes giving the painting a sense of movement and energy.

Prompt

Neon Impressionism, This image is a vibrant, abstract painting of a persons profile. The background is a bright yellow, and the figure is painted with thick, expressive brushstrokes in a variety of colors including blue, green, orange, pink, and purple. The persons hair is depicted with dark, textured strokes, and the facial features are highlighted with contrasting colors. The overall style is reminiscent of expressionism, with bold colors and dynamic, energetic brushwork. The painting captures a sense of movement and emotion, with the colors blending and overlapping in a lively, abstract manner.

Prompt

Neon Impression, This image is a vibrant and colorful painting of a tree. The tree is depicted using thick, textured brushstrokes, characteristic of the impasto technique. The trunk of the tree is dark and thick, with a rich blue hue, and it branches out into a canopy of leaves painted in a myriad of bright colors. The leaves are composed of various shades of blue, green, yellow, pink, and orange, creating a lively and dynamic appearance. The background of the painting is a warm, bright yellow, which contrasts beautifully with the cool colors of the leaves. The ground beneath the tree is painted with splashes of pink, blue, and green, adding to the overall sense of vibrancy and energy. The painting exudes a sense of joy and creativity, with the use of bold and contrasting colors. The overall effect is one of a lively, colorful landscape, with the tree standing out as the central focal point.

Prompt

Neon Impressionism, This image is a vibrant and dynamic painting of a surfer riding a wave. The artwork is characterized by its use of bright, bold colors and expressive brushstrokes, giving it a lively and energetic feel. The surfer is depicted in mid-action, crouched low on the surfboard, with one arm extended forward and the other bent backward for balance. The background is a swirl of colors, predominantly yellows, oranges, and greens, which adds to the sense of motion and excitement. The surfboard is painted in a pinkish hue, contrasting with the colorful background and the surfers multicolored shorts. The overall composition captures the thrill and intensity of surfing, with the artists use of color and movement creating a visually captivating scene.

Model description for Neon-Impressionism-Flux

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	27 & 3400
Epoch	22	Save Every N Epochs	1

Labeling: januspro-en(natural language & English)

Total Images Used for Training : 46

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/Halftone-Recraft-Flux"
trigger_word = "Neon Impressionism"  
pipe.load_lora_weights(lora_repo)

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