You can deploy this model and start earning money today!

srmahapatra95/
Cyberpunk-Lora-Weights

Name: Cyberpunk-Lora-Weights
Rating: 2.5 (38 reviews)

Tongyi-MAI/Z-Image-Turbo

adapter

mit

Views

No views yet

This repository contains LoRA (Low-Rank Adaptation) weights designed to infuse images with a gritty, neon-drenched Cyberpunk aesthetic. These weights are fine-tuned specifically for the Z-Image-Turbo base model, allowing for high-quality, futuristic generations in very few steps.

🚀 Model Details

Developed by: srmahapatra95
Model Type: LoRA (Image-to-Image)
Base Model: Tongyi-MAI/Z-Image-Turbo
Training Dataset: CyberVerse (Kaggle)
License: MIT

🎨 Style Characteristics

Neon Dominance: Enhances cyan, magenta, and electric blue lighting.
Atmospheric Effects: Adds "wet look" textures, rain-slicked streets, and cinematic fog.
Futuristic Detail: Increases complexity in mechanical parts, cybernetics, and urban architecture.

🛠 Usage Instructions

1. Using with Diffusers

Since this is a LoRA for a Turbo model, you can achieve great results in just 1–4 steps.

python

import torch
from diffusers import AutoPipelineForImage2Image
from diffusers.utils import load_image

# Load the base Turbo model
pipe = AutoPipelineForImage2Image.from_pretrained(
    "Tongyi-MAI/Z-Image-Turbo", 
    torch_dtype=torch.float16, 
    variant="fp16"
).to("cuda")

# Load this Cyberpunk LoRA
pipe.load_lora_weights("srmahapatra95/Cyberpunk-Lora-Weights")

# Initial Image
init_image = load_image("[https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/img2img-init.png](https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/img2img-init.png)")

# Prompt
prompt = "cyberpunk style, futuristic city, neon lights, high tech, rain, 8k"

# Generate (Keep steps low for Turbo models)
image = pipe(prompt, image=init_image, strength=0.6, num_inference_steps=4).images[0]
image.save("cyberpunk_output.png")