# QWEN-IMAGE Model |nf4|+Abliterated Qwen2.5VL-7b
This repo contains a variant of QWEN's **[QWEN-IMAGE](https://huggingface.co/Qwen/Qwen-Image)**, the state-of-the-art generative model with extensive and (image/)text-to-image &/or instruction/control-editing capabilities.
To make these cutting edge capabilities more accessible to those constrained to low-end consumer-grade hardware, **we've quantized the DiT (Diffusion Transformer) component of Qwen-Image to the 4-bit NF4 format** using the Bits&Bytes toolkit.
This optimization was derived by us directly from the BF16 base model weights released on 08/04/2025, with no other mix-ins or modifications to the DiT component.
*NOTE: Install `bitsandbytes` prior to inference.*
**QWEN-IMAGE** is an open-weights customization-friendly frontier model released under the highly permissive Apache 2.0 license, welcoming unrestricted (within legal limits) commercial, experimental, artistic, academic, and other uses &/or modifications.
To help highlight horizons of possibility broadened by the **QWEN-IMAGE** release, our quantization is bundled with an "Abliterated" (aka de-censored) finetune of [Qwen2.5-VL 7B Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct), QWEN-IMAGE model's sole conditioning encoder (of prompts, instructions, input images, controls, etc), as well as a powerful Vision-Language-Model in its own right.
As such, our repo saddles a lean & prim NF4 DiT over the **[Qwen2.5-VL-7B-Abliterated-Caption-it](https://huggingface.co/prithivMLmods/Qwen2.5-VL-7B-Abliterated-Caption-it/tree/main)** by [Prithiv Sakthi](https://huggingface.co/prithivMLmods) (aka [prithivMLmods](https://github.com/prithivsakthiur)).
# NOTICE:
*Do not be alarmed by the file warning from the ClamAV automated checker.*
*It is a clear false positive.* *In assessing one of the typical Diffusers-adapted Safetensors shards (model weights), the checker reads:*
``The following viruses have been found: Pickle.Malware.SysAccess.sys.STACK_GLOBAL.UNOFFICIAL``
*However, a Safetensors by its sheer design can not contain suchlike inserts. You may confirm for yourself thru HF's built-in weight/index viewer.
So, to be sure, this repo does **not** contain any pickle checkpoints, or any other pickled data.*
# TEXT-TO-IMAGE PIPELINE EXAMPLE:
This repo is formatted for usage with Diffusers (0.35.0.dev0+) & Transformers libraries, vis-a-vis associated pipelines & model component classes, such as the defaults listed in `model_index.json` (in this repo's root folder).
*Sourced/adapted from [the original base model repo](https://huggingface.co/Qwen/Qwen-Image) by QWEN.*
**EDIT:
We've confronted some issues with using the below pipeline. Will update once reliable adjustments are confirmed.**
```python
from diffusers import DiffusionPipeline
import torch
import bitsandbytes
model_name = "AlekseyCalvin/QwenImage_nf4"
# Load the pipeline
if torch.cuda.is_available():
torch_dtype = torch.bfloat16
device = "cuda"
else:
torch_dtype = torch.float32
device = "cpu"
pipe = DiffusionPipeline.from_pretrained(model_name, torch_dtype=torch_dtype)
pipe = pipe.to(device)
positive_magic = [
"en": "Ultra HD, 4K, cinematic composition." # for english prompt,
"zh": "超清，4K，电影级构图" # for chinese prompt,
]
# Generate image
prompt = '''A coffee shop entrance features a chalkboard sign reading "Qwen Coffee 😊 $2 per cup," with a neon light beside it displaying "通义千问". Next to it hangs a poster showing a beautiful Chinese woman, and beneath the poster is written "π≈3.1415926-53589793-23846264-33832795-02384197". Ultra HD, 4K, cinematic composition'''
negative_prompt = " "
# Generate with different aspect ratios
aspect_ratios = {
"1:1": (1328, 1328),
"16:9": (1664, 928),
"9:16": (928, 1664),
"4:3": (1472, 1140),
"3:4": (1140, 1472)
}
width, height = aspect_ratios["16:9"]
image = pipe(
prompt=prompt + positive_magic["en"],
negative_prompt=negative_prompt,
width=width,
height=height,
num_inference_steps=50,
true_cfg_scale=4.0,
generator=torch.Generator(device="cuda").manual_seed(42)
).images[0]
image.save("example.png")
```
# SHOWCASES FROM THE QWEN TEAM:
# MORE INFO:
- Check out the [Technical Report](https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-Image/Qwen_Image.pdf) for QWEN-IMAGE, released by the Qwen team!
- Find source base model weights here at [huggingface](https://huggingface.co/Qwen/Qwen-Image) and at [Modelscope](https://modelscope.cn/models/Qwen/Qwen-Image).
## QWEN LINKS:
💜 Qwen Chat | 🤗 Hugging Face | 🤖 ModelScope | 📑 Tech Report | 📑 Blog
🖥️ Demo | 💬 WeChat (微信) | 🫨 Discord
