---
language:
  - en
  - zh  # 如果支持中英双语
license: apache-2.0  # 或 mit, other 等
base_model_relation: quantized
tags:
  - image-generation
  - multimodal
  - qwen
  - quantized
  - nunchaku
  - comfyui
library_name: diffusers  # 或 diffusers 如果兼容
pipeline_tag: text-to-image  # 或 image-to-image，根据实际
base_model: Qwen/Qwen-Image-2512  # 如果有上游模型，填 repo id
---

# QuantFunc

<div align="center" style="margin-top: 50px;">
  <img src="assets/logo.webp" width="300" alt="Logo">
</div>

<p align="center">
  🤗 <a href="https://huggingface.co/QuantFunc">Hugging Face</a> |
  🤖 <a href="https://www.modelscope.cn/profile/QuantFunc">ModelScope</a> |
  💬 <a href="#wechat">WeChat (微信)</a> |
  🎮 <a href="https://discord.gg/jCp9TpFWcn">Discord</a>
</p>


## Introduction

We are excited to share our latest model series based on <strong>nunchaku + qwen-image-2512</strong> quantization. These models are carefully optimized to maintain high-quality output while significantly improving inference speed and efficiency. <strong>All models are 100% compatible with the nunchaku-comfyui && lora plugin</strong> and can be used directly in ComfyUI.

## Gallery

<div align="center">

<table>
  <tr>
    <td align="center"><img src="assets/ComfyUI_00006_.png" alt="Result 6" style="max-width: 300px;"></td>
    <td align="center"><img src="assets/ComfyUI_00010_.png" alt="Result 3" style="max-width: 300px;"></td>
  </tr>
  <tr>
    <td align="center"><img src="assets/ComfyUI_00019_.png" alt="Result 1" style="max-width: 300px;"></td>
    <td align="center"><img src="assets/ComfyUI_00051_.png" alt="Result 2" style="max-width: 300px;"></td>
  </tr>
  <tr>
    <td align="center"><img src="assets/ComfyUI_00040_.png" alt="Result 4" style="max-width: 300px;"></td>
    <td align="center"><img src="assets/ComfyUI_00048_.png" alt="Result 5" style="max-width: 300px;"></td>
  </tr>
</table>

</div>

## Model Checkpoints

| Name | low_rank | Notes |
|:---|:---:|:---|
| nunchaku_qwen_image_2512_best_quality_fp4 | 256 | Best quality model, suitable for scenarios with extremely high quality requirements |
| nunchaku_qwen_image_2512_best_quality_int4 | 256 | Best quality model, suitable for scenarios with extremely high quality requirements |
| nunchaku_qwen_image_2512_ultimate_speed_int4 | 32 | Ultimate speed model, prioritizing inference speed |
| nunchaku_qwen_image_2512_ultimate_speed_fp4 | 32 | Ultimate speed model, prioritizing inference speed |
| nunchaku_qwen_image_2512_balance_int4 | 128 | Balanced model, achieving the best balance between quality and speed |
| nunchaku_qwen_image_2512_balance_fp4 | 128 | Balanced model, achieving the best balance between quality and speed |

## 4 steps workflow

Here’s a <a href="https://huggingface.co/QuantFunc/Nunchaku-Qwen-Image-2512/tree/main">workflow</a> example of integrating 4-step LoRA in ComfyUI. If you don’t need 4-step LoRA, simply remove the LoRA node.
<img src="assets/workflow_image.png" alt="work flow" style="max-width: 800px;">

## Coming Soon

If you encounter any issues during use, feel free to join our community for feedback:
- Join our [Discord server](https://discord.gg/jCp9TpFWcn)
- Scan the QR code below to join our WeChat group

We will add support for build in lora and qwen-image-edit-2511 in approximately one month.

<div align="center">
  <img src="assets/Wechat.jpg" alt="WeChat Group" style="max-width: 300px;">
</div>