High-Resolution Inpainting - Kontext Image Editing LoRA

Model Introduction

This LoRA model is trained based on the Kontext model and DiffSynth-Studio. After loading this model, you can input the instruction Outpaint the image. to perform image outpainting.

Model Results

||Example 1|Example 2|Example 3|

|-|-|-|-|

|Prompt|Outpaint the image.|Outpaint the image.| Outpaint the image.|

|Input||||

|Outpainted Image||||

||Example 4|Example 5|Example 6|

|-|-|-|-|

|Prompt|Outpaint the image. A chicken-headed man in suspenders is playing basketball.|Outpaint the image. A man in suspenders is playing basketball.| Outpaint the image. A chicken-headed man in suspenders is playing basketball with a white background. |

|Input||||

|Outpainted Image||||

Usage Instructions

This model is built upon the DiffSynth-Studio framework. Please install it first:

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

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-SuperOutpainting", 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-SuperOutpainting/model.safetensors", alpha=1)

image = Image.open("your_image.jpg")

image = pipe(

prompt="Outpaint the image.",

kontext_images=image,

embedded_guidance=2.5,

seed=0,

)

image.save("output.jpg")