flux ghibsky illustration

Introduction

Flux Ghibsky Illustration is a text-to-image model designed to create serene and enchanting landscapes. It combines Ghibli-inspired whimsical charm with atmospheric beauty akin to Makoto Shinkai, making it ideal for generating dreamy visuals.

Architecture

The model generates landscapes by blending serene and surreal elements with intricate details. This fusion results in enchanting scenes that are both whimsical and atmospheric, capturing the essence of its stylistic inspirations.

Training

• Trained Using: ostris/flux-dev-lora-trainer
• Number of Images: 35
• Trigger Word: GHIBSKY
• Auto-captioning: Enabled
• Auto-captioning Suffix: ", GHIBSKY style"
• Training Steps: 1000
• Learning Rate: 0.0004
• Batch Size: 1
• LoRA Rank: 16

Guide: Running Locally

1. Installation: Ensure you have the Hugging Face Diffusers library installed.
2. Model Loading:

   from diffusers import AutoPipelineForText2Image
   import torch
   pipeline = AutoPipelineForText2Image.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16).to('cuda')
   pipeline.load_lora_weights('aleksa-codes/flux-ghibsky-illustration', weight_name='lora.safetensors')
   image = pipeline('GHIBSKY style, a serene lakeside village with colorful houses and towering mountains under a dreamy sky').images[0]
   

3. Execution: Run the code on a machine with a CUDA-compatible GPU for optimal performance. Consider using cloud services like AWS, Google Cloud, or Azure to access powerful GPUs.

License

The model is distributed under the flux-dev-license. Please review the terms at FLUX License to ensure compliance.