Tabby - 本地化AI代码自动补全 - Windows10

参考：

https://github.com/TabbyML/tabby

安装winget，方便命令行安装git

微软商店，搜索winget，安装App Installer

PS C:\Windows\system32> winget install --id Git.Git -e --source winget
Found Git [Git.Git] Version 2.41.0.2
This application is licensed to you by its owner.
Microsoft is not responsible for, nor does it grant any licenses to, third-party packages.
Downloading https://github.com/git-for-windows/git/releases/download/v2.41.0.windows.2/Git-2.41.0.2-64-bit.exe
                                  0.00 B / 58.3 MB

或者直接下载git.exe图像化安装

Git - Downloads (git-scm.com)

Git Clone tabby项目代码到本地

git clone https://github.com/TabbyML/tabby.git

Windows 10虚拟化设置

控制面板，程序

打开Windows功能，勾选上Hyper-V和虚拟机平台

wsl(Windows Subsystem for Linux)更新

管理员权限打开Windows Power Shell

执行命令查看相关子系统

PS C:\Windows\system32> wsl --list --online
The following is a list of valid distributions that can be installed.
The default distribution is denoted by '*'.
Install using 'wsl --install -d '.

  NAME            FRIENDLY NAME
* Ubuntu          Ubuntu
  Debian          Debian GNU/Linux
  kali-linux      Kali Linux Rolling
  openSUSE-42     openSUSE Leap 42
  SLES-12         SUSE Linux Enterprise Server v12
  Ubuntu-16.04    Ubuntu 16.04 LTS
  Ubuntu-18.04    Ubuntu 18.04 LTS
  Ubuntu-20.04    Ubuntu 20.04 LTS

不指定参数默认安装第一个星号标记的Ubuntu

> wsl --install

 输出：

PS C:\Windows\system32> wsl --install
Installing: Windows Subsystem for Linux
Windows Subsystem for Linux has been installed.
Downloading: WSL Kernel
Installing: WSL Kernel
WSL Kernel has been installed.
Downloading: Ubuntu
The requested operation is successful. Changes will not be effective until the system is rebooted.

重启电脑

下载并安装Docker

Docker: Accelerated, Containerized Application Development

Docker Desktop打开运行 

 

Docker Run下载镜像

Docker | Tabby

 

本文选择了CPU，打开PowerShell，在clone下来的tabby根目录下，执行下面的命令

docker run -p 8080:8080 -v $HOME/.tabby:/data tabbyml/tabby serve --model TabbyML/SantaCoder-1B

输出

如果遇到如下问题

​
ERROR tabby::serve: crates/tabby/src/serve/mod.rs:116: Failed to fetch model due to 'error sending request for url (https://huggingface.co/TabbyML/SantaCoder-1B/resolve/main/ctranslate2/shared_vocabulary.txt): error trying to connect: Connection reset by peer (os error 104)', is 'TabbyML/SantaCoder-1B' a valid model id?

解决方法

手动下载模型 

TabbyML/SantaCoder-1B at main

 

git lfs install
git clone https://huggingface.co/TabbyML/SantaCoder-1B

 大文件也可以直接网页手动下载

最后将上述SantaCoder完整目录文件移动到

$HOME/.tabby/models/TabbyML/目录下

 然后再重新执行

docker run -p 8080:8080 -v $HOME/.tabby:/data tabbyml/tabby serve --model TabbyML/SantaCoder-1B

模型部署过程中

 部署过程中需要等待，没有log可以显示进度，可以通过查看容器Container的stats的网络流量蓝色的数据接收一直在增长，表示部署中

 部署成功

（时间取决设备性能）

2023-07-19 08:36:57 2023-07-19T08:36:56.481802Z  INFO tabby::serve: crates/tabby/src/serve/mod.rs:131: Listening at 0.0.0.0:8080

 

 检查是否部署成功

浏览器本地访问 http://localhost:8080/

 或者在网络中使用   http://PC_IP:8080/

下载并安装VS Code(Visual Studio Code)

direct download link.Download Visual Studio Code - Mac, Linux, Windowsdirect download link.

  

 打开VS Code，扩展种搜索Tabby并安装

Tabby的管理里面设置上面本地部署的Tabby服务

 VS Code右下角显示Tabby已经可以提供服务

 

 开始高效编码吧

输入注释或者编码时，会自动补全，即灰色斜体部分，如果接收就直接按Tab键，采纳建议，否则正常进行编码