---
title: Qwen-Image-Edit-2511-ICEdit-LoRA
canonical_url: "https://www.modelscope.cn/models/DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA"
md_url: "https://www.modelscope.cn/models/DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA.md"
repository: DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA
chinese_name: "In-Context Editing LoRA (Qwen-Image-Edit-2511)"
last_updated: 2025-12-29
license: "Apache License 2.0"
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - Qwen/Qwen-Image-Edit-2511
base_model_relation: finetune
parameters: 235.9M
tensor_type:
  - F32
library_name:
  - pytorch
  - safetensors
frameworks:
  - Pytorch
supports_inference: img2img
downloads: 2373
stars: 38
---

# Qwen-Image-Edit-2511-ICEdit-LoRA

> Qwen-Image-Edit-2511-ICEdit-LoRA - DiffSynth-Studio 在 ModelScope 开源的模型。In-Context Editing LoRA (Qwen-Image-Edit-2511)

DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA 是 ModelScope 魔搭社区上的 235.9M 参数text-to-image-synthesis模型，采用 Apache License 2.0 许可，基于 Qwen/Qwen-Image-Edit-2511 构建，并支持在线推理（img2img）。

- **Repository**: DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA
- **License**: Apache License 2.0
- **Tasks**: text-to-image-synthesis
- **Parameters**: 235.9M
- **Base model**: Qwen/Qwen-Image-Edit-2511
- **Online inference**: img2img
- **Downloads**: 2373
- **Stars**: 38
- **Last updated**: 2025-12-29

Source: https://www.modelscope.cn/models/DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA

---

# In-Context Editing LoRA (Qwen-Image-Edit-2511)

## 模型介绍

这是一个很有趣的模型，能够为 [Qwen-Image-Edit-2511](https://modelscope.cn/models/Qwen/Qwen-Image-Edit-2511) 提供 In-Context Editing 能力。你可以为模型输入三张图：图1，图2，图3，模型会自动将图1到图2的变化应用到图3。


更多关于训练策略和实现细节，欢迎查看我们的[技术博客](https://modelscope.cn/learn/4826)。

## 效果展示

图片素材来源：

* 输入图 1：https://modelscope.cn/aigc/imageGeneration?tab=advanced&imageId=18968195
* 输入图 2：由编辑模型执行单图编辑生成
* 输入图 3：https://modelscope.cn/aigc/imageGeneration?tab=advanced&imageId=18723032

提示词：`Edit image 3 based on the transformation from image 1 to image 2.`

负向提示词：`泛黄，AI感，不真实，丑陋，油腻的皮肤，异常的肢体，不协调的肢体`

* 样例 1：表情参考

|输入图1|输入图2|输入图3|输出图|
|-|-|-|-|
|![](./assets/image1_original.png)|![](./assets/image1_edit_1.png)|![](./assets/image2_original.png)|![](./assets/image2_edit_1.png)|

* 样例 2：风格迁移

|输入图1|输入图2|输入图3|输出图|
|-|-|-|-|
|![](./assets/image1_original.png)|![](./assets/image1_edit_2.png)|![](./assets/image2_original.png)|![](./assets/image2_edit_2.png)|

* 样例 3：增加实体

|输入图1|输入图2|输入图3|输出图|
|-|-|-|-|
|![](./assets/image1_original.png)|![](./assets/image1_edit_3.png)|![](./assets/image2_original.png)|![](./assets/image2_edit_3.png)|

* 样例 4：局部编辑

|输入图1|输入图2|输入图3|输出图|
|-|-|-|-|
|![](./assets/image1_original.png)|![](./assets/image1_edit_4.png)|![](./assets/image2_original.png)|![](./assets/image2_edit_4.png)|

## 推理代码

安装 [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio):

```shell
git clone https://github.com/modelscope/DiffSynth-Studio.git  
cd DiffSynth-Studio
pip install -e .
```

推理代码：

```python
from diffsynth.pipelines.qwen_image import QwenImagePipeline, ModelConfig
from modelscope import snapshot_download
from PIL import Image
import torch

# Load models
pipe = QwenImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="Qwen/Qwen-Image-Edit-2511", origin_file_pattern="transformer/diffusion_pytorch_model*.safetensors"),
        ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="text_encoder/model*.safetensors"),
        ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    processor_config=ModelConfig(model_id="Qwen/Qwen-Image-Edit", origin_file_pattern="processor/"),
)
lora = ModelConfig(
    model_id="DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA",
    origin_file_pattern="model.safetensors"
)
pipe.load_lora(pipe.dit, lora)

# Load images
snapshot_download(
    "DiffSynth-Studio/Qwen-Image-Edit-2511-ICEdit-LoRA",
    local_dir="./data",
    allow_file_pattern="assets/*"
)
edit_image = [
    Image.open("data/assets/image1_original.png"),
    Image.open("data/assets/image1_edit_1.png"),
    Image.open("data/assets/image2_original.png")
]
prompt = "Edit image 3 based on the transformation from image 1 to image 2."
negative_prompt = "泛黄，AI感，不真实，丑陋，油腻的皮肤，异常的肢体，不协调的肢体"

# Generate
image_4 = pipe(
    prompt=prompt, negative_prompt=negative_prompt,
    edit_image=edit_image,
    seed=1,
    num_inference_steps=50,
    height=1280,
    width=720,
    zero_cond_t=True,
)
image_4.save("image.png")
```