If you already have your preferred Python environment, just check if your Python version is greater than or equal to 3.9.
Warning
It's not recommended to download the latest Python version, as some third-party libraries may not have adapted to the newest version yet, leading to plugin dependency issues.
In this guide, we will download the latest version of the previous major release.
Note
I personally dislike Conda installing its default environment on the system, so I won't explain how to use Conda here.
For an Anaconda virtual environment, refer to: nailongremove Deployment Guide
Visit the official Python website https://www.python.org, hover over "Downloads", and then click on "All Releases".
Scroll down to the "Looking for a specific release" section to find the version you want to download.
In the image above, you can see that the latest Python version is 3.13.0, so we'll look for the latest 3.12 version to download.
Click on the version, then scroll down to find the Windows installer (64-bit) for download. (Most computers today use X86 architecture with a 64-bit OS; if you have a different architecture, download the corresponding installer).
Once the download is complete, open the installer (ignore that this is the installation interface for version 3.11), make sure to check "Add python.exe to PATH"! Then click "Install Now" to start the installation.
If you want to change the installation path, you can choose "Customize Installation" and modify the path, but make sure the installation path does not contain any Chinese characters.
Open a terminal (press Win + R, type powershell, and press Enter), and run the following command to verify if Python is installed successfully:
python -V
On Linux, we use pyenv to manage Python versions.
To install, run the following command in the terminal:
curl https://pyenv.run | bashThen run the following commands to add the pyenv environment variable configurations to your shell configuration (if you're not using bash, replace the file name accordingly):
echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.bashrc
echo '[[ -d $PYENV_ROOT/bin ]] && export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.bashrc
echo 'eval "$(pyenv init -)"' >> ~/.bashrcReload the shell configuration:
source ~/.bashrcNow you should be able to use pyenv:
pyenv -v
Follow the instructions in the Wiki's Suggested build environment section to install the necessary dependencies based on your distribution. For example, for Ubuntu/Debian/Linux Mint:
Once the dependencies are installed, you can install Python 3.12 by running:
pyenv install 3.12Wait for it to finish, then set Python 3.12 as the default version for your shell:
pyenv global 3.12Finally, verify the installation:
python -V (Ignore the fact that the Python version here is 3.11)
Here, we use my developed nb-cli-plugin-bootstrap to create a NoneBot2 project.
Choose a directory where you'd like to place your NoneBot2 project. It's best if the path does not contain Chinese characters.
Windows users can open the terminal by clicking on the address bar in the File Explorer and typing powershell (ignore that I typed pwsh in the image), as shown below:
First, install nb-cli and nb-cli-plugin-bootstrap by running:
pip install nb-cli nb-cli-plugin-bootstrapAfter installation, run the following command to create the project:
nb bsThe project name will be the folder name where the project is created.
When choosing the adapter, use the up and down keys to select the "OneBot V11" adapter (if you're connecting to a wild QQ bot), and then press the space bar to select that adapter.
Be sure to press space to select the adapter you want to install! Once selected, the circle on the left will turn solid, and the item will light up—make sure to distinguish it!
Fill in the remaining options according to your preferences. If you don't want to customize, just keep pressing Enter to use the default settings.
It's recommended to change the port number to avoid conflicts with already occupied ports that might prevent NoneBot from starting.
Some settings can be modified later via the .env.prod file, which can be opened with any text editor for detailed information.
You will also need to fill in plugin configurations in this file.
After installation, it should look like this:
Next, switch the working directory to your newly created project directory by running the following command in the terminal:
Alternatively, you can open the project folder in File Explorer, click the address bar, type powershell, and press Enter to open a new terminal window.
cd 项目名Now, install the plugin by running:
nb plugin install nonebot-plugin-nailongremoveNote
The plugin installed this way only supports CPU-based AI model inference.
Configuring for GPU operation can be quite complex and is only recommended for advanced users. Please check the README for more details.
The installation process looks like this:
To start NoneBot, run the following command in the terminal:
nb runOn Windows, you can also double-click #启动.bat,
On Linux, you can run ./#run.sh.
NoneBot, start!
If everything goes smoothly without any errors, you should see the following:
If you're connecting to another chat platform (and installed other adapters in Step 2), skip this section and refer to the documentation for the corresponding platform adapter. Check the Adapter Store.
Choose one of the following protocol implementations that you prefer to start using.
A third-party NTQQ protocol implementation.
Go to the Actions page of the Lagrange repository, filter the Lagrange.OneBot Build workflow, then select the master branch (Click here).
(This step is for manually opening the page. If you accessed it through the above link, you can skip this step.)
Click the first checked item in the list on the right, then scroll down to find the "Artifacts" section, and download the build artifact for your machine's architecture.
If you cannot download the files from the list below, please log into your GitHub account.
How to identify your machine's architecture (from NoneBotX Community Docs) (Click to expand)
- For general personal computers or Windows Server, try the
win-x64version first; - For personal computers/servers running a Linux distribution, try the
linux-x64version first; - For ARM devices like Raspberry Pi or smartphones, try
linux-arm64first; if it doesn't work, trylinux-arm; - For Apple M-series chip compute
For Windows, X86 architecture CPU, 64-bit systems (which are typically used by most home computers and Windows cloud servers), download the following:
After the download is complete, open the compressed file, choose a directory where you'd like to extract the executable files, and then double-click to run it. Once you see the prompt like the one below, simply close it.
Open the appsettings.json file generated by Lagrange in the same directory.
Replace the 0 after Uin with your Bot's QQ number. Leave the password field blank and use the QR code login, as shown below.
Next, modify the configuration to enable a reverse WebSocket connection to NoneBot. Refer to the image below and remember to change the Port value from 8080 to the port number you previously configured.
Then, run Lagrange again, log in to your Bot's QQ account using the mobile phone QR code.
After that, you should see the message indicating that NoneBot is connected to your Bot and is able to receive messages.
NapCat is a protocol implementation based on QQNT, requiring the QQNT client to be installed. However, it can be run in headless mode (without a graphical user interface).
Follow the installation steps in the NapCat Documentation.
Once installed, modify the configuration to enable reverse WebSocket connections to NoneBot. Choose one of the following methods:
Find the WebUI address from the NapCat console, as shown in the box in the image below (both are the same, just different interface designs). Copy the address and open it in your browser.
Then, follow the steps in the image below to add the network configuration.
Locate NapCat's configuration file and refer to the configuration example below. Remember to change the 8080 value to the port number you previously configured.
{
// ...
"wsReverse": {
"enable": true,
"urls": ["ws://127.0.0.1:8080/onebot/v11/ws"]
}
// ...
}You don’t need to repeat the deployment steps. Just follow Section 4 to start your NoneBot, and then directly start the protocol endpoint you configured in Section 5.
Refer to the NoneBot Documentation on DotEnv Configuration for the file format, and modify the .env.prod file accordingly.
Example: If I want to modify the NAILONG_MODEL configuration to 1, I would modify the .env.prod file like this:
NAILONG_MODEL=1
Example of multi-line configuration items:
NAILONG_MODEL1_SCORE='
{
"nailong": 0.75,
"htgt": 0.75
}
'
NAILONG_TIP='
{
"nailong": "This group forbids sending 'nailong'!",
"htgt": "Suspected 'nailong' corpse chunk"
}
'
Open the command line in the Bot project directory and execute the following command to update the plugin:
nb plugin update nonebot-plugin-nailongremoveTo be added.
For other issues not listed, please visit the Issues section to search for solutions.