Qwen-Image-2.1-Turbo.py

from diffsynth.pipelines.qwen_image_21 import QwenImage21Pipeline, ModelConfig
import torch
from PIL import Image

vram_config = {
    "offload_dtype": "disk",
    "offload_device": "disk",
    "onload_dtype": "disk",
    "onload_device": "disk",
    "preparing_dtype": torch.bfloat16,
    "preparing_device": "cuda",
    "computation_dtype": torch.bfloat16,
    "computation_device": "cuda",
}
pipe = QwenImage21Pipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="Qwen/Qwen-Image-2.1-Turbo", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors", **vram_config),
        ModelConfig(model_id="Qwen/Qwen-Image-2.1", origin_file_pattern="text_encoder/model*.safetensors", **vram_config),
        ModelConfig(model_id="Qwen/Qwen-Image-2.1", origin_file_pattern="vae/diffusion_pytorch_model*.safetensors", **vram_config),
    ],
    processor_config=ModelConfig(model_id="Qwen/Qwen-Image-2.1", origin_file_pattern="processor/"),
    vram_limit=torch.cuda.mem_get_info("cuda")[1] / (1024 ** 3) - 0.5,
)
sigmas = [1.0, 0.978453, 0.95418, 0.926626, 0.89508, 0.845148, 0.704534, 0.414568]

# Text-to-Image, the output is an RGBA image
prompt = "Flat anime-style illustration, a girl with long black hair, wearing a JK uniform."
image = pipe(prompt, seed=0, sigmas=sigmas)
image.save("image1.png")

prompt = "Flat anime-style illustration, a sunny and cheerful high school girl."
image_2 = pipe(prompt=prompt, seed=0, sigmas=sigmas)
image_2.save("image2.png")

# Image Editing, the generated RGBA image is fed back as the condition
prompt = "Generate a group photo of these two characters."
edit_image = [Image.open("image1.png"), Image.open("image2.png")]
image_3 = pipe(prompt, edit_image=edit_image, seed=1, sigmas=sigmas)
image_3.save("image3.png")