We present In-Context Edit, a novel approach that achieves state-of-the-art instruction-based editing using just 0.5% of the training data and 1% of the parameters required by prior SOTA methods. The first row illustrates a series of multi-turn edits, executed with high precision, while the second and third rows highlight diverse, visually impressive single-turn editing results from our method.
:open_book: For more visual results, go checkout our
# To Do List
- [x] Inference Code
- [ ] Inference-time Scaling with VLM
- [x] Pretrained Weights
- [ ] More Inference Demos
- [x] Gradio demo
- [ ] Comfy UI demo
- [ ] Training Code
# News
- **[2025/4/30]** π₯ We release the inference code and [pretrained weights](https://huggingface.co/sanaka87/ICEdit-MoE-LoRA/tree/main) on Huggingface π€!
- **[2025/4/30]** π₯ We release the [paper](https://arxiv.org/abs/2504.20690) on arXiv!
- **[2025/4/29]** We release the [project page](https://river-zhang.github.io/ICEdit-gh-pages/) and demo video! Codes will be made available in next week~ Happy Labor Day!
# Installation
## Conda environment setup
```bash
conda create -n icedit python=3.10
conda activate icedit
pip install -r requirements.txt
```
## Download pretrained weights
If you can connect to Huggingface, you don't need to download the weights. Otherwise, you need to download the weights to local.
- [Flux.1-fill-dev](https://huggingface.co/black-forest-labs/flux.1-fill-dev).
- [ICEdit-MoE-LoRA](https://huggingface.co/sanaka87/ICEdit-MoE-LoRA).
## Inference in bash (w/o VLM Inference-time Scaling)
Now you can have a try!
> Our model can **only edit images with a width of 512 pixels** (there is no restriction on the height). If you pass in an image with a width other than 512 pixels, the model will automatically resize it to 512 pixels.
> If you found the model failed to generate the expected results, please try to change the `--seed` parameter. Inference-time Scaling with VLM can help much to improve the results.
```bash
python scripts/inference.py --image assets/girl.png \
--instruction "Make her hair dark green and her clothes checked." \
--seed 42 \
```
Editing a 512Γ768 image requires 35 GB of GPU memory. If you need to run on a system with 24 GB of GPU memory (for example, an NVIDIA RTX3090), you can add the `--enable-model-cpu-offload` parameter.
```bash
python scripts/inference.py --image assets/girl.png \
--instruction "Make her hair dark green and her clothes checked." \
--enable-model-cpu-offload
```
If you have downloaded the pretrained weights locally, please pass the parameters during inference, as in:
```bash
python scripts/inference.py --image assets/girl.png \
--instruction "Make her hair dark green and her clothes checked." \
--flux-path /path/to/flux.1-fill-dev \
--lora-path /path/to/ICEdit-MoE-LoRA
```
## Inference in Gradio Demo
We provide a gradio demo for you to edit images in a more user-friendly way. You can run the following command to start the demo.
```bash
python scripts/gradio_demo.py --port 7860
```
Like the inference script, if you want to run the demo on a system with 24 GB of GPU memory, you can add the `--enable-model-cpu-offload` parameter. And if you have downloaded the pretrained weights locally, please pass the parameters during inference, as in:
```bash
python scripts/gradio_demo.py --port 7860 \
--flux-path /path/to/flux.1-fill-dev (optional) \
--lora-path /path/to/ICEdit-MoE-LoRA (optional) \
--enable-model-cpu-offload (optional) \
```
Then you can open the link in your browser to edit images.
### π¨ Enjoy your editing!
# Comparison with Commercial Models
Compared with commercial models such as Gemini and GPT-4o, our methods are comparable to and even superior to these commercial models in terms of character ID preservation and instruction following. We are more open-source than them, with lower costs, faster speed (it takes about 9 seconds to process one image), and powerful performance.
# Bibtex
If this work is helpful for your research, please consider citing the following BibTeX entry.
```
@misc{zhang2025ICEdit,
title={In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformer},
author={Zechuan Zhang and Ji Xie and Yu Lu and Zongxin Yang and Yi Yang},
year={2025},
eprint={2504.20690},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2504.20690},
}
```