README.md

metadata

license: apache-2.0
datasets:
  - svjack/pokemon-blip-captions-en-zh
language:
  - zh
library_name: diffusers
tags:
  - stable-diffusion
  - lora

Model Card for souljoy/sd-pokemon-model-lora-zh

Installation

pip install -U diffusers
pip install transformers

Usage

from diffusers import StableDiffusionPipeline
import torch

pipe = StableDiffusionPipeline.from_pretrained("IDEA-CCNL/Taiyi-Stable-Diffusion-1B-Anime-Chinese-v0.1", torch_dtype=torch.float16)
model_path = "souljoy/sd-pokemon-model-lora-zh"

pipe.unet.load_attn_procs(model_path)
pipe.to("cuda")
pipe.safety_checker = lambda images, clip_input: (images, False)

prompt = "粉色的蝴蝶,小精灵,卡通"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]
image

prompt = "可爱的狗,小精灵,卡通"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]
image

prompt = "漂亮的猫,小精灵,卡通"
image = pipe(prompt, num_inference_steps=50, guidance_scale=7.5).images[0]
image