最近配了人生中第一台主机，上了个4070ti super显卡，玩3A游戏眩晕，只能拿这个卡跑跑大模型来玩玩了。先装个stable-diffusion玩玩
安装python
- 安装pyenv
ubuntu 24.04自带的版本是3.12，sd-webui要求是3.10或者3.11。使用pyenv安装管理python版本
➜ ~ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 24.04.1 LTS
Release: 24.04
Codename: noble
➜ ~ python3 --version
Python 3.12.3
# 安装pyenv
curl https://pyenv.run | bash
# 安装 python 3.11.10
pyenv install 3.11.10
# 切换到3.11.10版本
pyenv global 3.11.10
curl [https://pyenv.run](https://pyenv.run) | bash安装不上可以修改一下自己的hosts，查一下域名对应的ip配置到hosts中
185.199.111.133 raw.githubusercontent.com
安装失败的话可能是缺少依赖
➜ ~ pyenv install 3.11.10
Downloading Python-3.11.10.tar.xz...
-> https://www.python.org/ftp/python/3.11.10/Python-3.11.10.tar.xz
Installing Python-3.11.10...
BUILD FAILED (Ubuntu 24.04 using python-build 20180424)
Inspect or clean up the working tree at /tmp/python-build.20250105203930.3197
Results logged to /tmp/python-build.20250105203930.3197.log
Last 10 log lines:
checking for pkg-config... no
checking for --enable-universalsdk... no
checking for --with-universal-archs... no
checking MACHDEP... "linux"
checking for gcc... no
checking for cc... no
checking for cl.exe... no
configure: error: in `/tmp/python-build.20250105203930.3197/Python-3.11.10':
configure: error: no acceptable C compiler found in $PATH
See `config.log' for more details
安装相关依赖
- 基本开发工具和依赖
apt install -y build-essential libssl-dev zlib1g-dev libbz2-dev \
libreadline-dev libsqlite3-dev wget curl llvm libncurses5-dev libncursesw5-dev \
xz-utils tk-dev libffi-dev liblzma-dev python3-dev python3-venv
- 安装gcc
apt install gcc - 安装make
apt install make
安装cude
参考cuda官网
wget https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/cuda-wsl-ubuntu.pin
sudo mv cuda-wsl-ubuntu.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/12.6.3/local_installers/cuda-repo-wsl-ubuntu-12-6-local_12.6.3-1_amd64.deb
sudo dpkg -i cuda-repo-wsl-ubuntu-12-6-local_12.6.3-1_amd64.deb
sudo cp /var/cuda-repo-wsl-ubuntu-12-6-local/cuda-*-keyring.gpg /usr/share/keyrings/
sudo apt-get update
sudo apt-get -y install cuda-toolkit-12-6
安装完成后**<font style="color:rgb(44, 62, 80);">nvidia-smi</font>**查看安装效果
➜ download nvidia-smi
Sun Jan 5 21:13:30 2025
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 565.72 Driver Version: 566.14 CUDA Version: 12.7 |
|-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 4070 ... On | 00000000:01:00.0 On | N/A |
| 0% 36C P0 46W / 295W | 1614MiB / 16376MiB | 1% Default |
| | | N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| No running processes found |
+-----------------------------------------------------------------------------------------+
安装sd-weiui
git clone https://github.com/AUTOMATIC1111/stable-diffusion-webui
## 执行
webui.sh
不允许root启动，去webui.sh 修改一下代码 允许root启动 搜索root 自己改吧。
执行./webui.sh 报错
Cannot activate python venv, aborting…
#在stable-diffusion-webui目录下执行
python3 -c 'import venv'
python3 -m venv venv
Cannot locate TCMalloc (improves CPU memory usage)
sudo apt install libgoogle-perftools4 libtcmalloc-minimal4 -y
