stable diffusion xl + wrong Loraを使う

brew install cmake protobuf rust python@3.10 git wget

A woman with an umbrella walks along a street surrounded by New York City buildings. Numerous yellow taxis crisscross the street in the rain.

Drawing only Asians, avoiding Caucasians and African Americans.

from diffusers import DiffusionPipeline
import torch

pipe = DiffusionPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0",
    torch_dtype=torch.float16,
    use_safetensors=True,
    variant="fp16"
    )
pipe.to("cuda")

# 画像生成の実行
prompt = "cute cat ear maid"
image = pipe(prompt=prompt).images[0]

# 保存と表示
image.save("output.png")
image

OutOfMemoryError: CUDA out of memory. Tried to allocate 512.00 MiB (GPU 0; 15.77 GiB total capacity; 12.83 GiB already allocated; 484.12 MiB free; 13.81 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF```

# 画像生成の実行
prompt = "A woman with an umbrella walks along a street surrounded by New York City buildings. Numerous yellow taxis crisscross the street in the rain."
image = pipe(prompt=prompt).images[0]

# 保存と表示
image.save("output.png")
image

A woman with an umbrella walks along a street surrounded by New York City buildings. Numerous yellow taxis crisscross the street in the rain. --v 5.2