Bioluminescence Style

Introduction

The Bioluminescence Style LoRA is a specialized model designed to apply a dazzling, glowing light effect to your images. Trained on the FLUX.1-Kontext-dev base model, this LoRA excels at creating a distinct visual style reminiscent of natural bioluminescence.

This model allows creators to infuse their artwork with vibrant, luminous edges and highlights. Its primary capability is to transform standard images by adding a radiant, glowing aesthetic, making it ideal for projects that require a touch of fantasy, sci-fi, or magical realism. The effect is triggered by a specific prompt, allowing for targeted application of the style.

How to Use

To generate images with the Bioluminescence Style, you need to load this LoRA with the FLUX.1-Kontext-dev base model. The provided documentation recommends using the diffusers library.

Step-by-Step Inference

Follow this Python code example to run inference:

import torch
from diffusers import FluxKontextPipeline
from diffusers.utils import load_image

# Load the base pipeline
pipe = FluxKontextPipeline.from_pretrained("black-forest-labs/FLUX.1-Kontext-dev", torch_dtype=torch.bfloat16)

# Load the LoRA weights
pipe.load_lora_weights("Shakker-Labs/FLUX.1-Kontext-dev-LoRA-Bioluminescence-Style", weight_name="FLUX-kontext-lora-bioluminescence-style.safetensors")

# Fuse the LoRA and move to GPU
pipe.fuse_lora(lora_scale=1.0)
pipe.to("cuda")

# Prepare your input image and prompt
input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cat.png")
prompt = "bioluminescence efficiency"

# Generate the image
image = pipe(
    image=input_image,
    prompt=prompt, 
    num_inference_steps=24, 
    guidance_scale=2.5,
).images[0]

# Save the output
image.save(f"example.png")

Trigger Word

To activate the style, you must include bioluminescence efficiency in your prompt.

Recommended Settings

The example provided in the documentation uses the following settings, which can be a good starting point:

• LoRA Scale: lora_scale=1.0
• Inference Steps: num_inference_steps=24
• Guidance Scale: guidance_scale=2.5

Use Cases

The primary use case for this LoRA is to apply a "dazzling light effect style" to images. It is designed to add glowing and luminous properties to subjects, enhancing them with a bioluminescent quality.

You can use simple or descriptive prompts to achieve the effect. Based on the provided examples, you can use:

• The main trigger word: bioluminescence efficiency
• A descriptive sentence: The edges of the character's body emit a dazzling light effect.

Examples

The model is capable of producing striking visual effects, as demonstrated in the showcase gallery on the model page. By using the trigger word bioluminescence efficiency, the LoRA applies a radiant glow to the subject of an image. The effect can be seen on the edges and contours of objects and characters, creating a vibrant and eye-catching aesthetic. The example images show how this effect can be applied to various subjects to give them a magical or futuristic appearance.

FAQ

Q: What is the trigger word for this LoRA?

A: You should use bioluminescence efficiency in your prompt to trigger the image generation style.

Q: What base model is required to use this LoRA?

A: This LoRA was trained on and is intended to be used with the black-forest-labs/FLUX.1-Kontext-dev base model.

Q: Can I use this model for commercial purposes?

A: This model is released under the flux-1-dev-non-commercial-license. Please refer to the full license text for detailed terms and restrictions on use.

Q: What LoRA scale is recommended?

A: The provided inference example uses a lora_scale of 1.0. You can experiment with this value, but this is the documented setting.

Q: Who trained this model?

A: The model was trained by the user zooey and is released under permission.

Technical Details

• Base Model: black-forest-labs/FLUX.1-Kontext-dev
• Model Type: flux-dev LoRA
• File Format: The model is provided in .safetensors format.
• Training: The model was trained by user zooey on the FLUX.1-Kontext-dev model. The instance prompt used during training was bioluminescence efficiency.
• Compatibility: The provided code example demonstrates usage with the diffusers library and the FluxKontextPipeline.