WeShop_UI_1.0.0 / README.md
WeShop's picture
Upload README.md
dd56651 verified

WeShop-UI

Welcome to Weshop_UI_1.0.0! This software is an open-source solution derived from WeShop.AI, which specialize in generating images using AI. We've extracted some business/professional functionalities, removed proprietary elements, and are gradually releasing parts of our services to the public. Our goal is to foster innovation and collaboration in AI-powered image creation while maintaining our core business offerings. WeShop UI features multi-task management, and a simple and user-friendly interface, perfect for beginners.

Introduction

Weshop_UI_1.0.0 is an open-source solution that brings some functionalities of weshop.ai to your local operation. Weshop_UI_1.0.0 leverages Stable Diffusion Webui as its core inference engine, enabling you to deploy WeShop.AI functionalities locally and generate high-quality AI images without internet or cloud reliance.

News

  • 2024.07 We officially launched WeShop UI

Open-source Plan

Currently, we've open sourced web and Electron frontend, changes made for Automatic1111 Stable Diffusion Webui. Our plan regarding key functionalities is listed below:

  • Keep upgrading Webui with the latest one on https://github.com/AUTOMATIC1111/stable-diffusion-webui

  • Open source the front end part of this project

  • Refactor and open source the backend part of this project. This project uses compiled executables currently.

  • Support ComfyUI simultaneously.

  • Support multiple-GPU inference scheduling.

  • Gradually adapt workflows/features from WeShop.AI

  • More functions to come and feel free to leave you feedbacks!

Features

  1. Seamlessly integrated SD WebUI functionalities, supporting SDXL model and SDXL's Controlnet

  2. Enable shared graphics card utilization for enhanced performance

  3. Harness the power of robust task management functionalities

  4. Unlock batch task execution, interrupt capabilities, and execute single tasks on a large scale for unparalleled efficiency

How to use WeShop AI Locally

alt text

Click the "Start" button on the bottom right to launch the tool. Depending on the computer's specification, it takes some time to launch. alt text alt text

This website will pop up after the software is launched. Select the type of task desired to be executed. Currently, functionalities include text-to-image (txt2img) and image-to-image (img2img). alt text

Select desired parameters for image generations. alt text alt text

Further refine generations by selecting desired ControlNet and AfterDetailer parameters. Click the "Generate" button on bottom right to start generation process. alt text

Resulting images will shortly be displayed after generation process is finished. Click "Edit again" button on the top right to adjust parameters or generate again. alt text

For any bugs encountered, click the "BUG Report" button, then bug info will be sent. Alternatively, you can send feedback from our Github page https://github.com/weshopai/WeShop-UI

Requirements

  • Windows 7 or above system

  • 8GB VRAM Nvidia GPU

  • 16GB RAM

Download

You can download WeShop UI 1.0.0 from the following sources:

Huggingface: https://huggingface.co/WeShopAI/WeShop_UI_1.0.0

Steam: in progress

Installation & Usage

Unzip weshop-1.0.0.zip and weshop-model-pack.zip

Replace weshop/models with weshop-model-pack/models

Replace weshop/extensions/sd-webui-controlnet/models with weshop-model-pack/extensions/sd-webui-controlnet/models

Replace weshop/extensions/sd-webui-controlnet/annotator/downloads with weshop-model-pack/extensions/sd-webui-controlnet/annotator/downloads

Click weshop.exe to launch

For Steam users, find Weshop_AI in the market, then directly launch the software from Steam.

Gallery

alt text alt text alt text alt text alt text

Acknowledgements

This project makes use of the following open-source projects:

https://github.com/sgreben/http-file-server: A file server to store necessary local files created

https://github.com/pierrre/imageserver: A file server dedicated to store image files created

https://github.com/facebookresearch/segment-anything: Segment uploaded images to create masks

https://github.com/sczhou/CodeFormer: Upscale images