Skip to content
ModelVerifiedExternal
Browse files

Qwen Image Edit 2511 SDNQ Uint4 Svd R32

4 bit (UINT4 with SVD rank 32) quantization of Qwen/Qwen-Image-Edit-2511 using SDNQ.

Sourcehuggingface.co/Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32/tree/a285fceef1439d72d60533e83cb6d8921748a666
KindText encoderBase modelQwen/Qwen-Image-Edit-2511VersionRevision a285fcePublisher@Disty0Unratedgrade

Model source

Kind
Text encoder
File size
242 MB–5.05 GB
Base model
Qwen/Qwen-Image-Edit-2511
Version
vRevision a285fce
Updated
Oct 4, 2026
Publisher
@Disty0
Downloads
854
Tasks
Image edit, Image to image
Source
Hugging Face

Source description

4 bit (UINT4 with SVD rank 32) quantization of Qwen/Qwen-Image-Edit-2511 using SDNQ.

Usage:

pip install sdnq

import torch
import diffusers
from sdnq import SDNQConfig # import sdnq to register it into diffusers and transformers
from sdnq.common import use_torch_compile as triton_is_available
from sdnq.loader import apply_sdnq_options_to_model

pipe = diffusers.QwenImageEditPlusPipeline.from_pretrained("Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32", torch_dtype=torch.bfloat16)

# Enable INT8 MatMul for AMD, Intel ARC and Nvidia GPUs:
if triton_is_available and (torch.cuda.is_available() or torch.xpu.is_available()):
    pipe.transformer = apply_sdnq_options_to_model(pipe.transformer, use_quantized_matmul=True)
    pipe.text_encoder = apply_sdnq_options_to_model(pipe.text_encoder, use_quantized_matmul=True)
    # pipe.transformer = torch.compile(pipe.transformer) # optional for faster speeds

pipe.enable_model_cpu_offload()
pipe.set_progress_bar_config(disable=None)

image1 = Image.open("input1.png")
image2 = Image.open("input2.png")
prompt = "The magician bear is on the left, the alchemist bear is on the right, facing each other in the central park square."
inputs = {
    "image": [image1, image2],
    "prompt": prompt,
    "generator": torch.manual_seed(0),
    "true_cfg_scale": 4.0,
    "negative_prompt": " ",
    "num_inference_steps": 40,
    "guidance_scale": 1.0,
    "num_images_per_prompt": 1,
}
with torch.inference_mode():
    output = pipeline(**inputs)
    output_image = output.images[0]
    output_image.save("qwen-image-edit-2511-sdnq-uint4-svd-r32.png")

Requirements

Models

1

Sources

1 source
Hugging FacePrimaryVersion a285fceef1439d72d60533e83cb6d8921748a666
huggingface.co/Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32/tree/a285fceef1439d72d60533e83cb6d8921748a666

Verified Oct 4

Download files

Model artifacts

5 artifacts

text_encoder/model.safetensors

safetensors · 5.05 GB · SHA-256 397c575b427c…7054 · Hugging Face

Download file

transformer/diffusion_pytorch_model-00001-of-00003.safetensors

safetensors · 4.64 GB · SHA-256 f4d9935fadd5…3146 · Hugging Face

Download shard

transformer/diffusion_pytorch_model-00002-of-00003.safetensors

safetensors · 4.63 GB · SHA-256 b7d220f3db8f…748b · Hugging Face

Download shard

transformer/diffusion_pytorch_model-00003-of-00003.safetensors

safetensors · 1.52 GB · SHA-256 90933dd851d4…d77c · Hugging Face

Download shard

vae/diffusion_pytorch_model.safetensors

safetensors · 242 MB · SHA-256 0c8bc8b758c6…8344 · Hugging Face

Download file

Source excerpts

3 excerpts
ReadmeJul 11huggingface.co/Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32

--- license: apache-2.0 base_model: - Qwen/Qwen-Image-Edit-2511 base_model_relation: quantized library_name: diffusers tags: - sdnq - qwen_image - 4-bit --- 4 bit (UINT4 with SVD rank 32) quantization of [Qwen/Qwen-Image-Edit-2511](https://huggingface.co/Qwen/Qwen-Image-Edit-2511) using [SDNQ](https://github.com/Disty0/sdnq). Usage: ``` pip install sdnq ``` ```py import torch import diffusers from sdnq import SDNQConfig # import sdnq to register it into diffusers and transformers from sdnq.common import use_torch_compile as triton_is_available from sdnq.loader import apply_sdnq_options_to_model pipe = diffusers.QwenImageEditPlusPipeline.from_pretrained("Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32", torch_dtype=torch.bfloat16) # Enable INT8 MatMul for AMD, Intel ARC and Nvidia GPUs: if triton_is_available and (torch.cuda.is_available() or torch.xpu.is_available()): pipe.transformer = apply_sdnq_options_to_model(pipe.transformer, use_quantized_matmul=True) pipe.text_encoder = apply_sdnq_options_to_model(pipe.text_encoder, use_quantized_matmul=True) # pipe.transformer = torch.compile(pipe.transformer) # optional for faster speeds pipe.enable_model_cpu_offload() pipe.set_progress_bar_config(disable=None) image1 = Image.open("input1.png") image2 = Image.open("input2.png") prompt = "The magician bear is on the left, the alchemist bear is on the right, facing each other in the central park square." inputs = { "image": [image1, image2], "prompt": prompt, "generator": torch.manual_seed(0), "true_cfg_scale": 4.0, "negative_prompt": " ", "num_inference_steps": 40, "guidance_scale": 1.0, "num_images_per_prompt": 1, } with torch.inference_mode(): output = pipeline(**inputs) output_image = output.images[0] output_image.save("qwen-image-edit-2511-sdnq-uint4-svd-r32.png") ```

NoteJul 18huggingface.co/Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32

Source context: 56587 downloads · 8 likes · Pipeline image-to-image · Library diffusers · Repo Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32

NoteJul 14huggingface.co/Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32

Source context: 69899 downloads · 8 likes · Pipeline image-to-image · Library diffusers · Repo Disty0/Qwen-Image-Edit-2511-SDNQ-uint4-svd-r32