ComfyUI Extension: ComfyUI-Text-Translation

Authored by mo230761

Created

Updated

7 stars

This repository provides the official ComfyUI workflow for a/Insert Anything.

Custom Nodes (0)

    README

    This repository provides the official ComfyUI workflow for Insert Anything.

    🔧 Installation

    We offer two methods for installing InsertAnything-ComfyUI.

    Prerequisites

    Please first install ComfyUI with the following commands:

       git clone https://github.com/comfyanonymous/ComfyUI.git
       cd ComfyUI
       pip install -r requirements.txt
    

    ComfyUI-Manager

    1. Install ComfyUI-Manager with the following commands:
       cd custom_nodes
       git clone https://github.com/ltdrdata/ComfyUI-Manager
    
    1. Launch ComfyUI:
       cd ..	# Make sure you are in the ComfyUI root directory
       python main.py
    
    1. Open the ComfyUI Manager, select Install via Git URL, input the URL:
       https://github.com/mo230761/InsertAnything-ComfyUI-official.git
    
    1. Move the InsertAnything.json file into your local ComfyUI's user/default/workflows directory.

    Manual Installation

    1. Clone this repository into the custom_nodes directory inside ComfyUI:
       cd custom_nodes
       git clone https://github.com/mo230761/InsertAnything-ComfyUI-official.git
    
    1. Move the InsertAnything.json file into your local ComfyUI's user/default/workflows directory

    2. Launch ComfyUI:

       cd ..	# Make sure you are in the ComfyUI root directory
       python main.py
    

    🚀 Usage Examples

    Using this workflow, the source image and ref_image are cropped and scaled to a standard size of 768*768 to generate the image.

    The effect is shown in the figure below.

    insert_anything

    The Insert Anything LoRA still performs well at larger resolutions. Therefore, to avoid image quality loss caused by processing images to 768*768, you can also use this workflow to only crop the image for generation. In this case, you can set the image resolution more freely.

    The effect is shown in the figure below.

    insert_anything_noscaling