README.md

---
license: apache-2.0
---

# High-Resolution Fix - Kontext Image Editing LoRA

## Model Introduction

This LoRA model is trained based on the [Kontext](https://www.modelscope.cn/models/black-forest-labs/FLUX.1-Kontext-dev) model and [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio). After using this model, you can input the instruction `Improve the clarity.` to enhance the image clarity.

## Model Results

||Example 1|Example 2|Example 3|
|-|-|-|-|
|Original|![](./assets/image_0.jpg)|![](./assets/image_6.jpg)|![](./assets/image_14.jpg)|
|Generated|![](./assets/image_0_fix.jpg)|![](./assets/image_6_fix.jpg)|![](./assets/image_14_fix.jpg)|

||Example 4|Example 5|Example 6 (Yes, it can handle normal images)|
|-|-|-|-|
|Original|![](./assets/image_1.jpg)|![](./assets/image_3.jpg)|![](./assets/image_2.jpg)|
|Generated|![](./assets/image_1_fix.jpg)|![](./assets/image_3_fix.jpg)|![](./assets/image_2_fix.jpg)|

## Usage Instructions

This model is trained using the [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio/tree/main/examples/flux) framework. Please install it first:

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

```python
import torch
from diffsynth.pipelines.flux_image_new import FluxImagePipeline, ModelConfig
from PIL import Image
from modelscope import snapshot_download

snapshot_download("DiffSynth-Studio/FLUX.1-Kontext-dev-lora-highresfix", cache_dir="./models")
pipe = FluxImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="black-forest-labs/FLUX.1-Kontext-dev", origin_file_pattern="flux1-kontext-dev.safetensors"),
        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder/model.safetensors"),
        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder_2/"),
        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="ae.safetensors"),
    ],
)
pipe.load_lora(pipe.dit, "models/DiffSynth-Studio/FLUX.1-Kontext-dev-lora-highresfix/model.safetensors", alpha=1)

image = Image.open("your_image.jpg")
image = pipe(
    prompt="Improve the clarity.",
    kontext_images=image,
    embedded_guidance=2.5,
    seed=0,
)
image.save("output.jpg")
```