import torch
import diffusers
from sdnq import SDNQConfig # import sdnq to register it into diffusers and transformers

pipe = diffusers.ChromaPipeline.from_pretrained("Disty0/Chroma1-HD-SDNQ-uint4-svd-r32", torch_dtype=torch.bfloat16)
pipe.enable_model_cpu_offload()

prompt = [
    "A high-fashion close-up portrait of a blonde woman in clear sunglasses. The image uses a bold teal and red color split for dramatic lighting. The background is a simple teal-green. The photo is sharp and well-composed, and is designed for viewing with anaglyph 3D glasses for optimal effect. It looks professionally done."
]
negative_prompt =  ["low quality, ugly, unfinished, out of focus, deformed, disfigure, blurry, smudged, restricted palette, flat colors"]

image = pipe(
    prompt=prompt,
    negative_prompt=negative_prompt,
    generator=torch.Generator("cpu").manual_seed(433),
    num_inference_steps=40,
    guidance_scale=3.0,
    num_images_per_prompt=1,
).images[0]
image.save("chroma-1-hd-sdnq-uint4-svd-r32.png")

Original BF16 vs SDNQ quantization comparison:

Quantization	Model Size	Visualization
Original BF16	17.8 GB	Original BF16
SDNQ UINT4	5.4 GB	SDNQ UINT4