sprodem/sdxl-naruto-lora

SDXL Naruto Style LoRA

This LoRA was trained using train_text_to_image_lora_sdxl.py on the dataset:

• sprodem/naruto-style-sdxl

Training Details

• Base model: stabilityai/stable-diffusion-xl-base-1.0
• VAE: madebyollin/sdxl-vae-fp16-fix
• Resolution: 1024×1024
• Rank: 64
• Mixed precision: fp16
• Optimizer: 8-bit Adam
• Steps: 2 (example)
• Validation prompt: "Naruto anime portrait style"

Usage

Load base SDXL:

pipe = StableDiffusionXLPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16).to("cuda")

Apply LoRA:

pipe.load_lora_weights("sprodem/sdxl-naruto-lora")

Generate an image:

image = pipe("Naruto anime portrait style").images[0]