Qwen image transformers doesn't currently support from_single_file (i.e. GGUFs)

[luke14free]

Currently it looks like qwen image doesn't support gguf since it's missing a conversion map/function from gguf to diffuser. I'd love to help creating it if you can point me in the right direction, I tried looking into it but without guidance it's above my paygrade :)

[DN6]

Hi @luke14free support for this is incoming (should be merged by tomorrow) 👍🏽

[sukitfup]

hero

[DN6]

Hi @luke14free Should be working now that #11684 was merged.

Snippet

from diffusers import QwenImageTransformer2DModel, GGUFQuantizationConfig, DiffusionPipeline
import torch

ckpt_path = "https://huggingface.co/city96/Qwen-Image-gguf/blob/main/qwen-image-Q2_K.gguf"
transformer = QwenImageTransformer2DModel.from_single_file(
    ckpt_path,
    quantization_config=GGUFQuantizationConfig(compute_dtype=torch.bfloat16),
    torch_dtype=torch.bfloat16,
    config="Qwen/Qwen-Image",
    subfolder="transformer",
)
pipe = DiffusionPipeline.from_pretrained("Qwen/Qwen-Image", transformer=transformer, torch_dtype=torch.bfloat16).to("cuda")
prompt = "stock photo of two people, a man and a woman, wearing lab coats writing on a white board with markers, the white board has text that reads 'The Diffusers library by Hugging Face makes it easy for developers to run image generation and inference using state-of-the-art diffusion models with just a few lines of code' with sloppy writing and traces clearly made by a human. The photo is taken from the side and has depth of field so some parts of the board looks blurred giving it a more professional look"

image = pipe(
    prompt=prompt,
    negative_prompt="negative_prompt",
    width=1024,
    height=1024,
    num_inference_steps=25,
    true_cfg_scale=4.0,
    generator=torch.manual_seed(0),
).images[0]
image.save("gguf_qwen.png")

[luke14free]

big love to you @DN6 works perfectly!