| --- |
| language: |
| - en |
| - zh |
| license: apache-2.0 |
| pipeline_tag: image-to-image |
| library_name: diffusers |
| arxiv: 2602.13344 |
| --- |
| |
| <p align="center"> |
| <img src="./assets/logo.png" width="600"/> |
| <p> |
| <p align="center" style="line-height: 1;"> |
| <a href="https://huggingface.co/FireRedTeam" target="_blank"><img alt="Hugging Face" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-FireRedTeam-ffc107?color=ffc107&logoColor=white" style="display: inline-block;"/></a> |
| <a href="https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.0" target="_blank"><img alt="Hugging Face Model" src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-FireRed--Image--Edit--1.0-red" style="display: inline-block;"/></a> |
| <a href="https://huggingface.co/spaces/FireRedTeam/FireRed-Image-Edit-1.0" target="_blank"><img alt="Demo" src="https://img.shields.io/badge/%F0%9F%92%BB%20Demo-FireRed--Image--Edit--1.0-red" style="display: inline-block;"/></a> |
| </p> |
| |
| <p align="center" style="line-height: 1;"> |
| ๐ค <a href="https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.0">HuggingFace</a> | |
| ๐ฅ๏ธ <a href="https://huggingface.co/spaces/FireRedTeam/FireRed-Image-Edit-1.0"> Demo</a> | |
| ๐ <a href="https://huggingface.co/papers/2602.13344">Technical Report</a> | |
| ๐ป <a href="https://github.com/FireRedTeam/FireRed-Image-Edit">Code</a> |
| </p> |
| <p align="center"> |
| <img src="./assets/teaser.png" width="800"/> |
| <p> |
| |
|
|
| ## ๐ฅ FireRed-Image-Edit |
| **FireRed-Image-Edit** is a general-purpose image editing model that delivers high-fidelity and consistent editing across a wide range of scenarios. It is introduced in the paper [FireRed-Image-Edit-1.0 Technical Report](https://huggingface.co/papers/2602.13344). |
|
|
| ## โจ Key Features |
| - **Strong Editing Performance**: FireRed-Image-Edit delivers leading open-source results with accurate instruction following, high image quality, and consistent visual coherence. |
| - **Native Editing Capability**: Built directly from text-to-image foundation model and endowed with editing capabilities. |
| - **Text Style Preservation**: Maintains text styles with high fidelity, achieving performance comparable to closed-source solutions. |
| - **Photo Restoration**: High-quality old photo restoration and enhancement. |
| - **Multi-Image Editing**: Flexible editing of multiple images such as virtual try-on. |
|
|
|
|
| ## ๐ฐ News |
| - 2026.02.14: We released FireRed-Image-Edit-1.0 model weights. Check more details in the [Model Zoo](#-model-zoo) section. |
| - 2026.02.10: We released the [Technical Report](https://huggingface.co/papers/2602.13344) of FireRed-Image-Edit-1.0. |
|
|
| ## ๐จ Showcase |
| Some real outputs produced by FireRed-Image-Edit across general editing. |
| <p align="center"> |
| <img src="./assets/showcase.png" width="800"/> |
| <p> |
| |
| ## ๐๏ธ Model Zoo |
|
|
| <div style="overflow-x: auto; margin-bottom: 16px;"> |
| <table style="border-collapse: collapse; width: 100%;"> |
| <thead> |
| <tr> |
| <th style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de; background-color: #f6f8fa;">Models</th> |
| <th style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de; background-color: #f6f8fa;">Task</th> |
| <th style="padding: 8px; border: 1px solid #d0d7de; background-color: #f6f8fa;">Description</th> |
| <th style="padding: 8px; border: 1px solid #d0d7de; background-color: #f6f8fa;">Download Link</th> |
| </tr> |
| </thead> |
| <tbody> |
| <tr> |
| <td style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de;">FireRed-Image-Edit-1.0</td> |
| <td style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de;">Image-Editing</td> |
| <td style="padding: 8px; border: 1px solid #d0d7de;">General-purpose image editing model</td> |
| <td style="padding: 8px; border: 1px solid #d0d7de;"> |
| <span style="white-space: nowrap;">๐ค <a href="https://huggingface.co/FireRedTeam/FireRed-Image-Edit-1.0">HuggingFace</a></span> |
| </td> |
| </tr> |
| <tr> |
| <td style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de;">FireRed-Image-Edit-1.0-Distilled</td> |
| <td style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de;">Image-Editing</td> |
| <td style="padding: 8px; border: 1px solid #d0d7de;">Distilled version of FireRed-Image-Edit-1.0 for faster inference</td> |
| <td style="padding: 8px; border: 1px solid #d0d7de;"> |
| <span style="white-space: nowrap;">To be released</span> |
| </td> |
| </tr> |
| <tr> |
| <td style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de;">FireRed-Image</td> |
| <td style="white-space: nowrap; padding: 8px; border: 1px solid #d0d7de;">Text-to-Image</td> |
| <td style="padding: 8px; border: 1px solid #d0d7de;">High-quality text-to-image generation model</td> |
| <td style="padding: 8px; border: 1px solid #d0d7de;"> |
| <span style="white-space: nowrap;">To be released</span> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| |
| ## ๐๏ธ Model Architecture |
| <p align="center"> |
| <img src="./assets/architecture.png" width="800"/> |
| <p> |
| |
| ## โก๏ธ Quick Start |
|
|
| 1. Install the latest version of diffusers |
| ``` |
| pip install git+https://github.com/huggingface/diffusers |
| ``` |
| 2. Use the following code snippets to generate or edit images. |
| ``` |
| python inference.py \ |
| --input_image ./examples/edit_example.png \ |
| --prompt "ๅจไนฆๆฌๅฐ้ขPython็ไธๆน๏ผๆทปๅ ไธ่ก่ฑๆๆๅญ2nd Edition" \ |
| --output_image output_edit.png \ |
| --seed 43 |
| ``` |
|
|
| ## ๐ Benchmark |
| To better validate the capabilities of our model, we propose a benchmark called REDEdit-Bench. Our main goal is to build more diverse scenarios and editing instructions that better align with human language, enabling a more comprehensive evaluation of current editing models. We collected over 3,000 images from the internet, and after careful expert-designed selection, we constructed 1,673 bilingual (ChineseโEnglish) editing pairs across 15 categories. |
|
|
| ### Inference and Evaluation Code |
| We provide the inference and evaluation code for REDEdit-Bench. Please refer to the [redbench_infer.py](./src/tools/redbench_infer.py) and [redbench_eval.py](./src/tools/redbench_eval.py) scripts in the `src/tools` directory for more details. |
|
|
| ### Results on ImgEdit |
|
|
| <table style="border-collapse:collapse; width:100%; font-family:system-ui, -apple-system, Segoe UI, Roboto, Arial, sans-serif; font-size:14px;"> |
| <thead> |
| <tr> |
| <th style="border:1px solid #ddd; padding:8px; text-align:left;">Model</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Overall โ</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Add</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Adjust</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Extract</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Replace</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Remove</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">BG</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Style</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Hybrid</th> |
| <th style="border:1px solid #ddd; padding:8px; text-align:center;">Action</th> |
| </tr> |
| </thead> |
| <tbody><tr> |
| <td colspan="11" style="border:1px solid #ddd; padding:8px; font-weight:700; background:#f7f7f7;"> |
| ๐น Proprietary Models |
| </td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Nano-Banana</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.29</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.62</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.41</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>3.68</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.34</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.39</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.40</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.18</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>3.72</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.83</b></td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Seedream4.0</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.30</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.33</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.38</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>3.89</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.65</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.57</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.35</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.22</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>3.71</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.61</td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Seedream4.5</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.32</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.57</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.65</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">2.97</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.66</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.46</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.37</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.92</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>3.71</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.56</td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Nano-Banana-Pro</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.37</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.44</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.62</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.42</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.60</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.63</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.32</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.97</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.64</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.69</u></td> |
| </tr><tr> |
| <td colspan="11" style="border:1px solid #ddd; padding:8px; font-weight:700; background:#f7f7f7;" > |
| ๐น Open-source Models |
| </td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">FLUX.1 Kontext [Dev]</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.71</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.99</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.88</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">2.19</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.27</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.13</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.98</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.51</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.23</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.18</td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Step1X-Edit-v1.2</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.95</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.91</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.04</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">2.68</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.48</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.26</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.90</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.82</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.23</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.22</td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Qwen-Image-Edit-2509</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.31</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.34</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.27</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.42</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.73</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.36</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.37</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.91</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.56</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.80</td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">FLUX.2 [Dev]</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.35</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.50</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.18</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.83</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.65</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.65</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.31</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.88</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.46</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.70</td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">LongCat-Image-Edit</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.45</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.44</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.53</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.83</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.80</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.60</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.33</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.92</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">3.75</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.82</b></td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;">Qwen-Image-Edit-2511</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.51</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.54</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.57</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.13</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.70</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.46</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.36</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.89</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.16</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.81</u></td> |
| </tr><tr> |
| <td style="border:1px solid #ddd; padding:8px; text-align:left;"><b>FireRed-Image-Edit</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.56</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.55</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.66</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.34</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.75</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.58</td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.45</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><b>4.97</b></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;"><u>4.07</u></td> |
| <td style="border:1px solid #ddd; padding:8px; text-align:center;">4.71</td> |
| </tr></tbody> |
| </table> |
| |
| ## ๐ License Agreement |
|
|
| The code and the weights of FireRed-Image-Edit are licensed under Apache 2.0. |
|
|
|
|
| ## ๐ TODO: |
| - [x] Release FireRed-Image-Edit-1.0 model. |
| - [ ] Release REDEdit-Bench, a comprehensive benchmark for image editing evaluation. |
| - [ ] Release FireRed-Image-Edit-1.0-Distilled model, a distilled version of FireRed-Image-Edit-1.0 for few-step generation. |
| - [ ] Release FireRed-Image model, a text-to-image generative model. |
|
|
|
|
| ## ๐๏ธ Citation |
|
|
| We kindly encourage citation of our work if you find it useful. |
|
|
| ```bibtex |
| @article{firered2026rededit, |
| title={FireRed-Image-Edit: A General-Purpose Image Editing Model}, |
| author={Super Intelligence Team}, |
| year={2026}, |
| archivePrefix={arXiv}, |
| primaryClass={cs.CV}, |
| url={https://huggingface.co/papers/2602.13344}, |
| } |
| ``` |
|
|
| ## ๐ค Acknowledgements |
|
|
| We would like to thank the developers of the amazing open-source projects, including [Qwen-Image](https://github.com/QwenLM/Qwen-Image), [Diffusers](https://github.com/huggingface/diffusers) and [HuggingFace](https://huggingface.co). |
|
|
| ## โญ Star History |
|
|
| [](https://www.star-history.com/#FireRedTeam/FireRed-Image-Edit&type=date&legend=top-left) |