保姆级教程！如何安装运行stable -diffusion-webui？

2024-05-3009:43:06AI绘画与短视频剪辑Comments416 views字数 9745阅读模式

▌锅头导读文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

最近，锅头花了一周时间，断断续续调试解决安装过程中遇到的问题，终于完成了stable-diffusion-webui的启动并成功生成了stable-diffusion-webui。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

在这里做个笔记，也顺便分享给大家，希望能帮助少踩一些坑。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

下面，跟锅头一起学习stable-difffusion-webui吧。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌前期准备

硬件设备文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

华为Matebook 16s文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

处理器 13th Gen Intel(R) Core(TM) i9-13900H 2.60 GHz文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

机带 RAM 32.0 GB (31.7 GB 可用)文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

系统类型 64 位操作系统, 基于 x64 的处理器文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

笔和触控为 10 触摸点提供触控支持文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

版本 Windows 11 家庭中文版
版本 23H2
安装日期 ‎2023/‎10/‎12
操作系统版本 22631.3593
体验 Windows Feature Experience Pack 1000.22700.1003.0文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

操作步骤文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

安装Python。
安装Git。
安装Rust。
克隆stable-diffusion-webui到本地电脑。
下载stable-diffusion模型。
运行stable-diffusion-webui。

注意事项文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

锅头的电脑是Intel核显（Intel(R) Iris(R) Xe Graphics），所以选择的stable-diffusion-webui，需要用支持 Intel 显卡的版本，即OpenVINO版本。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

OpenVINO下载链接：https://github.com/openvinotoolkit/stable-diffusion-webui文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

Python版本：Python 3.10.11文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

PS：关于Python 版本，最开始下载的是Python 3.12.3版本，但是安装依赖包时提示不支持，提示要下载Python3.8.x版本。接着锅头又下载安装了3.8.x版本，发现在安装依赖包时又提示不支持。最后下载了Python3.10.11才能把必要的依赖包安装完成。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌安装Python

参考文章【AI学习】保姆级教程，手把手如何在电脑上安装Python（Windows用户），完成Python安装。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

PS：需要注意的是Python版本，锅头下载最新的3.12.3版本，安装依赖包时提示不支持，建议选择3.10.x版本。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌安装Git文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

参考文章【AI学习】保姆级教程，手把手如何在电脑上安装Git（Windows用户），完成Git安装。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

安装Git是为了更方便将stable-diffusion-webui克隆至本地电脑上。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌安装Rust文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

参考文章【AI学习】保姆级教程！手把手教你如何下载安装Rust？（AI协助解决问题），完成Rust安装。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

安装Rust是为了解决构建tokenizers包时出现错误问题。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌克隆CPU版本的stable-diffusion-webui文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入本地电脑的D盘的stable-diffusion文件夹下，新建文件夹openvino。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

以管理员身份运行命令提示符窗口，进入openvino文件夹路径下。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

输入命令行如下：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

用Git克隆https://github.com/openvinotoolkit/stable-diffusion-webui.git项目。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

输入命令行：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

git clone https://github.com/openvinotoolkit/stable-diffusion-webui.git

克隆项目成功，看到openvino文件夹多了stable-diffusion-webui文件夹项目。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

到这里，完成了CPU版本的stable-diffusion-webui项目克隆。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌下载Stable Diffusion模型文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

在浏览器输入https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/v1-5-pruned-emaonly.safetensors，按回车进入stable diffusion模型下载页面。点击红框中的【download】文字，下载模型。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

下载中，有4G，下载需要一些时间。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

下载好模型后，将其复制粘贴至当前项目的models/Stable-diffusion文件夹路径下。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

D:\stable-diffusion\openvino\stable-diffusion-webui\models\Stable-diffusion文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

PS：后续需要启用新的模型也是放入该路径下即可。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

更多模型下载可进入下面两个网站下载：文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

https://civitai.com/文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

https://www.liblib.art/文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌创建Python虚拟环境文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

以管理员身份运行命令提示符，进入命令提示符窗口。
文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入d盘，输入命令行：d:文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入stable-diffusion-webui文件夹，输入命令行：cd D:\stable-diffusion\openvino\stable-diffusion-webui文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

创建名称为sd的虚拟环境，输入命令行：python -m venv sd文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌激活Python虚拟环境文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

以管理员身份运行命令提示符，进入命令提示符窗口。
文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入d盘，输入命令行：d:文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入文件夹D:\stable-diffusion\openvino\stable-diffusion-webui\venv\Scripts路径下，输入命令行：cd D:\stable-diffusion\openvino\stable-diffusion-webui\venv\Scripts。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入文件夹后，输入命令行：activate，激活环境。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入sd虚拟环境。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

到这里，咱们成功激活sd虚拟环境，接下来可以在这里启动stable-diffusion-webui啦。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

▌启动stable-diffusion-webui

升级pip版本

在sd虚拟环境中，安装依赖包前，先把pip升级至最新版本，避免后续因pip版本导致安装失败的问题。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

命令行：python.exe -m pip install --upgrade pip文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

启动stable-diffusion-webui

第一次启动stable-diffusion-webui，缺少的依赖包会自动下载并安装。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

进入(sd) D:\stable-diffusion\openvino\stable-diffusion-webui>路径下，输入启动项目命令行：py launch.py --skip-torch-cuda-test --precision full --no-half文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

PS：由于我的电脑没有 NVIDIA 显卡，需要在运行launch.py时增加“跳过 PyTorch CUDA 测试，使用全精度，禁止半精度”，命令行如下：py launch.py --skip-torch-cuda-test --precision full --no-half文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

首次启动下载安装依赖包耗时比较久，锅头网速1M/S左右，约耗时1小时20分钟。文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html

(sd) D:\stable-diffusion\openvino\stable-diffusion-webui\sd\Scripts>python.exe -m pip install --upgrade pipRequirement already satisfied: pip in d:\stable-diffusion\openvino\stable-diffusion-webui\sd\lib\site-packages (23.0.1)Collecting pip Downloading pip-24.0-py3-none-any.whl (2.1 MB) ---------------------------------------- 2.1/2.1 MB 940.0 kB/s eta 0:00:00Installing collected packages: pip Attempting uninstall: pip Found existing installation: pip 23.0.1 Uninstalling pip-23.0.1: Successfully uninstalled pip-23.0.1Successfully installed pip-24.0(sd) D:\stable-diffusion\openvino\stable-diffusion-webui\sd\Scripts>py launch.py --skip-torch-cuda-test --precision full --no-halfD:\Program Files\python310\python.exe: can't open file 'D:\\stable-diffusion\\openvino\\stable-diffusion-webui\\sd\\Scripts\\launch.py': [Errno 2] No such file or directory(sd) D:\stable-diffusion\openvino\stable-diffusion-webui\sd\Scripts>cd ..(sd) D:\stable-diffusion\openvino\stable-diffusion-webui\sd>cd ..(sd) D:\stable-diffusion\openvino\stable-diffusion-webui>py launch.py --skip-torch-cuda-test --precision full --no-halffatal: No names found, cannot describe anything.Python 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]Version: 1.6.0Commit hash: e5a634da06c62d72dbdc764b16c65ef3408aa588Installing torch and torchvisionLooking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cu118Collecting torch==2.0.1 Downloading https://download.pytorch.org/whl/cu118/torch-2.0.1%2Bcu118-cp310-cp310-win_amd64.whl (2619.1 MB) ---------------------------------------- 2.6/2.6 GB 581.4 kB/s eta 0:00:00Collecting torchvision==0.15.2 Downloading https://download.pytorch.org/whl/cu118/torchvision-0.15.2%2Bcu118-cp310-cp310-win_amd64.whl (4.9 MB) ---------------------------------------- 4.9/4.9 MB 792.0 kB/s eta 0:00:00Collecting filelock (from torch==2.0.1) Downloading filelock-3.14.0-py3-none-any.whl.metadata (2.8 kB)Collecting typing-extensions (from torch==2.0.1) Downloading typing_extensions-4.11.0-py3-none-any.whl.metadata (3.0 kB)Collecting sympy (from torch==2.0.1) Downloading https://download.pytorch.org/whl/sympy-1.12-py3-none-any.whl (5.7 MB) ---------------------------------------- 5.7/5.7 MB 531.7 kB/s eta 0:00:00Collecting networkx (from torch==2.0.1) Downloading networkx-3.3-py3-none-any.whl.metadata (5.1 kB)Collecting jinja2 (from torch==2.0.1) Downloading jinja2-3.1.4-py3-none-any.whl.metadata (2.6 kB)Collecting numpy (from torchvision==0.15.2) Downloading numpy-1.26.4-cp310-cp310-win_amd64.whl.metadata (61 kB) ---------------------------------------- 61.0/61.0 kB 1.6 MB/s eta 0:00:00Collecting requests (from torchvision==0.15.2) Downloading requests-2.32.2-py3-none-any.whl.metadata (4.6 kB)Collecting pillow!=8.3.*,>=5.3.0 (from torchvision==0.15.2) Downloading pillow-10.3.0-cp310-cp310-win_amd64.whl.metadata (9.4 kB)Collecting MarkupSafe>=2.0 (from jinja2->torch==2.0.1) Downloading https://download.pytorch.org/whl/MarkupSafe-2.1.5-cp310-cp310-win_amd64.whl (17 kB)Collecting charset-normalizer<4,>=2 (from requests->torchvision==0.15.2) Downloading charset_normalizer-3.3.2-cp310-cp310-win_amd64.whl.metadata (34 kB)Collecting idna<4,>=2.5 (from requests->torchvision==0.15.2) Downloading idna-3.7-py3-none-any.whl.metadata (9.9 kB)Collecting urllib3<3,>=1.21.1 (from requests->torchvision==0.15.2) Downloading urllib3-2.2.1-py3-none-any.whl.metadata (6.4 kB)Collecting certifi>=2017.4.17 (from requests->torchvision==0.15.2) Downloading certifi-2024.2.2-py3-none-any.whl.metadata (2.2 kB)Collecting mpmath>=0.19 (from sympy->torch==2.0.1) Downloading https://download.pytorch.org/whl/mpmath-1.3.0-py3-none-any.whl (536 kB) ---------------------------------------- 536.2/536.2 kB 560.7 kB/s eta 0:00:00Downloading pillow-10.3.0-cp310-cp310-win_amd64.whl (2.5 MB) ---------------------------------------- 2.5/2.5 MB 938.3 kB/s eta 0:00:00Downloading filelock-3.14.0-py3-none-any.whl (12 kB)Downloading jinja2-3.1.4-py3-none-any.whl (133 kB) ---------------------------------------- 133.3/133.3 kB 984.2 kB/s eta 0:00:00Downloading networkx-3.3-py3-none-any.whl (1.7 MB) ---------------------------------------- 1.7/1.7 MB 820.2 kB/s eta 0:00:00Downloading numpy-1.26.4-cp310-cp310-win_amd64.whl (15.8 MB) ---------------------------------------- 15.8/15.8 MB 904.1 kB/s eta 0:00:00Downloading requests-2.32.2-py3-none-any.whl (63 kB) ---------------------------------------- 63.9/63.9 kB 383.3 kB/s eta 0:00:00Downloading typing_extensions-4.11.0-py3-none-any.whl (34 kB)Downloading certifi-2024.2.2-py3-none-any.whl (163 kB) ---------------------------------------- 163.8/163.8 kB 1.1 MB/s eta 0:00:00Downloading charset_normalizer-3.3.2-cp310-cp310-win_amd64.whl (100 kB) ---------------------------------------- 100.3/100.3 kB 1.2 MB/s eta 0:00:00Downloading idna-3.7-py3-none-any.whl (66 kB) ---------------------------------------- 66.8/66.8 kB 898.3 kB/s eta 0:00:00Downloading urllib3-2.2.1-py3-none-any.whl (121 kB) ---------------------------------------- 121.1/121.1 kB 786.0 kB/s eta 0:00:00Installing collected packages: mpmath, urllib3, typing-extensions, sympy, pillow, numpy, networkx, MarkupSafe, idna, filelock, charset-normalizer, certifi, requests, jinja2, torch, torchvisionSuccessfully installed MarkupSafe-2.1.5 certifi-2024.2.2 charset-normalizer-3.3.2 filelock-3.14.0 idna-3.7 jinja2-3.1.4 mpmath-1.3.0 networkx-3.3 numpy-1.26.4 pillow-10.3.0 requests-2.32.2 sympy-1.12 torch-2.0.1+cu118 torchvision-0.15.2+cu118 typing-extensions-4.11.0 urllib3-2.2.1Installing clipInstalling open_clipInstalling requirements for CodeFormerInstalling requirementsLaunching Web UI with arguments: --skip-torch-cuda-test --precision full --no-halfno module 'xformers'. Processing without...no module 'xformers'. Processing without...No module 'xformers'. Proceeding without it.Warning: caught exception 'Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx', memory monitor disabledCalculating sha256 for D:\stable-diffusion\openvino\stable-diffusion-webui\models\Stable-diffusion\stableDiffusionXL_v10Roundeyes.safetensors: Running on local URL: http://127.0.0.1:7860To create a public link, set `share=True` in `launch()`.Startup time: 4703.9s (prepare environment: 4677.7s, import torch: 10.2s, import gradio: 3.8s, setup paths: 4.1s, initialize shared: 0.6s, other imports: 3.2s, setup codeformer: 0.2s, load scripts: 2.8s, create ui: 1.0s, gradio launch: 0.2s).8744a0dc449420be1ef1cc51588fe4388b395ce1c17166e67edccfa7d4a3c40aLoading weights [8744a0dc44] from D:\stable-diffusion\openvino\stable-diffusion-webui\models\Stable-diffusion\stableDiffusionXL_v10Roundeyes.safetensorsCreating model from config: D:\stable-diffusion\openvino\stable-diffusion-webui\repositories\generative-models\configs\inference\sd_xl_base.yamlApplying attention optimization: InvokeAI... done.Model loaded in 16.5s (calculate hash: 8.4s, load weights from disk: 0.9s, create model: 0.7s, apply weights to model: 3.3s, apply float(): 2.3s, calculate empty prompt: 0.9s).文章源自菜鸟学院-https://www.cainiaoxueyuan.com/jianji/63738.html