Qwen-Image.py

# Ref: https://huggingface.co/Qwen/Qwen-Image/discussions/7
from diffusers import DiffusionPipeline, PipelineQuantizationConfig
from diffusers import BitsAndBytesConfig
import torch

quant_config = PipelineQuantizationConfig(
    quant_mapping={
        "transformer": BitsAndBytesConfig(
            load_in_4bit=True,
            bnb_4bit_compute_dtype=torch.bfloat16,
            bnb_4bit_quant_type="nf4",
            llm_int8_skip_modules=[
                "time_text_embed",
                "img_in",
                "norm_out",
                "proj_out",
                "img_mod",
                "txt_mod",
            ],
        ),
    }
)

pipe = DiffusionPipeline.from_pretrained(
    "Qwen/Qwen-Image",
    quantization_config=quant_config,
    torch_dtype=torch.bfloat16
)
pipe.enable_model_cpu_offload()

prompt = "Japanese anime style, a cat holding a sign that says hello world"
negative_prompt="3d, cg, photo"
image = pipe(
    prompt=prompt,
    negative_prompt=negative_prompt,
    num_inference_steps=50,
).images[0]
image.save("qwenimage_nf4.png")

生成物