# Stable-Diffusion-Webui 指南 [TOC] # 簡介 **Stable Diffusion**是一種基於AI的圖像生成技術,它通過將圖像逐步從高度隨機的噪聲狀態進行重建,最終生成一個清晰的圖像。這種技術可以用於生成各種風格和主題的圖像,如風景、人物、建築等,並能夠根據特定的指令生成具有特定特徵和風格的圖像,甚至催生了新職業ai詠唱師。**Stable Diffusion**在AI圖像生成領域具有很高的潛力,並且因為其免費和易用的特點,受到許多創作者和設計師的青睞 ![](https://hackmd.io/_uploads/SJrFD0HVh.png =25%x25%)![](https://hackmd.io/_uploads/BJoFwRrN2.png =25%x25%)![](https://hackmd.io/_uploads/HkevtCSE2.png =25%x25%)![](https://hackmd.io/_uploads/HkEwtCrV3.png =25%x25%) **Stable Diffusion-Webui** 提供使用Stable Diffusion的GUI介面 ![Imgur](https://imgur.com/W06D4XB.png) https://github.com/AUTOMATIC1111/stable-diffusion-webui ## Automatic Installation 主要介紹如何在Windows與Linux上安裝Stable Diffusion-Webui ### Automatic Installation on Windows 1. Install Python 3.10.6 (Newer version of Python does not support torch), checking "Add Python to PATH". 2. Install git. 3. Download the stable-diffusion-webui repository, for example by running python : https://www.python.org/downloads/windows/ ![](https://hackmd.io/_uploads/BJnu3kUNn.png) > **紅色箭頭指向的地放記得打勾** **git** : https://git-scm.com/ ``` git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git ``` Run `webui-user.bat` from Windows Explorer as normal, non-administrator, user. ### Automatic Installation on Linux Install the dependencies: ```sudo apt install wget git python3 python3-venv``` Navigate to the directory you would like the webui to be installed and execute the following command: ``` git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui.git ``` Run `webui.sh` ![](https://hackmd.io/_uploads/HkJvRkLN2.png) ## 介面介紹 ![](https://hackmd.io/_uploads/ryxLitkUV3.png) ## 各種風格模型下載 Stable Diffusion最方便的地方就在很多神人開發了很多各式各樣的風格模型讓大家可以使用,當然自己也可以訓練出自己的風格模型,而目前最方便的地方就是[Civitai](https://civitai.com/)網站了 [Civitai](https://civitai.com/) 下載後將 ckpt檔案 放至 models/Stable-diffusion 目錄下,就可以在模型清單看到該模型 ![](https://imgur.com/nz13bOO.png) ## Demo [Demo_1](http://140.113.24.139:7860/) [Demo_2](http://140.113.24.139:7861/) [Demo_3](http://140.113.24.139:7862/) [Demo_4](http://140.113.24.139:7863/) ## 目前提供的模型 台灣食物 : https://civitai.com/models/21079/common-taiwanese-food-or RPG角色:https://civitai.com/models/1116/rpg Prompt: https://huggingface.co/Anashel/rpg/resolve/main/RPG-V4-Model-Download/RPG-Guide-v4.pdf Ink punk : https://civitai.com/models/1087/inkpunk-diffusion 素描anime: https://civitai.com/models/16014/anime-lineart-manga-like-style 建築:https://civitai.com/models/10063/isometric-future IU:https://civitai.com/models/11722/iu Lyriel:https://civitai.com/models/22922/lyriel ## References Stable diffusion(github):https://github.com/CompVis/stable-diffusion https://vocus.cc/article/63eee139fd8978000188d7d8 https://mnya.tw/cc/word/1887.html https://gooptions.cc/stable-diffusion%E6%95%99%E5%AD%B8/ https://jet3c.com/ai-2/ ## ChatGPT生成咒語 https://gooptions.cc/ai%E7%B9%AA%E5%9C%96%E5%92%92%E8%AA%9Eprompt%E7%94%9F%E6%88%90%E5%99%A8/