README.md

metadata

base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: photo of a billie777 person
tags:
  - text-to-image
  - diffusers
  - autotrain
inference: true

Billie Eillish face trained with DreamBooth

It's properly work only in 35mm lenses photo

Example Prompt

portrait photo of a billie777 person wear cyberpunk jacket,4k,35mm lenses

Cyberpunk Art

portrait photo of a billie777 person wear cyberpunk jacket, neon light, cyberpunk art, retrofuturism, glowing neon, neon, retrowave, 4k, 35mm lenses

output:

Usage

from diffusers import DiffusionPipeline, StableDiffusionXLImg2ImgPipeline
import torch

 prj_path = "/content/sdxl-billie-eillish-neo"
 model = "stabilityai/stable-diffusion-xl-base-1.0"
 pipe = DiffusionPipeline.from_pretrained(
     model,
     torch_dtype=torch.float16,
)
 pipe.to("cuda")
 pipe.load_lora_weights(prj_path, weight_name="pytorch_lora_weights.safetensors")

 prompt = "portrait photo of a billie777 person wear cyberpunk jacket, neon light, cyberpunk art, retrofuturism, glowing neon, neon, retrowave, 4k, 35mm lenses"
 negative_prompt = "extra arms,extra fingers,extra legs,extra limbs,fused fingers,long neck,low quality,lowres,malformed limbs,missing arms,missing legs,mutated hands,poorly drawn hands,too many fingers,worst quality"

 seed = 42
 generator = torch.Generator("cuda").manual_seed(seed)
 image = pipe(prompt=prompt, negative_prompt=negative_prompt, generator=generator).images[0]
 image.save(f"generated_image.png")

Best Negative Prompt

extra arms,extra fingers,extra legs,extra limbs,fused fingers,long neck,low quality,lowres,malformed limbs,missing arms,missing legs,mutated hands,poorly drawn hands,too many fingers,worst quality

Example Generated images