Discord Bot Guide

Why is he so annoyed?

Overview

This guide will help you set up a Discord bot that uses a local language model to respond to messages. The bot is built using Python and the discord.py library, and it leverages models from Hugging Face.

Created by Nano Chances. Tutorial video TBD.

File Structure

File	Description
index.py	Main bot code.
requirements.txt	List of required Python libraries.
.env	Environment variables (e.g., Discord bot token).
readme.md	This guide.
venv/	Virtual environment folder (created during setup).

Setup Instructions

Download Python 3.10.6 from python.org.

Make sure to check the box that says Add Python to PATH during installation.

You can find required libraries in requirements.txt.

Discord Dev Portal

1. Go to the Discord Developer Portal.
2. Click on "New Application" and give it a name.
3. Navigate to the "Bot" tab and reset the token. Copy this token and save it in the .env file as DISCORD_BOT_TOKEN=your_token_here.
4. Scroll down to "Privileged Gateway Intents" and enable the "Message Content Intent".
5. Go to the "OAuth2" tab, then "URL Generator". Under "Scopes", select bot. Under "Bot Permissions", select Send Messages. Copy the generated URL and open it in your browser to invite the bot to your server.
6. Run the bot using python index.py in your terminal and check if it comes online in your server.

Opening a Terminal

You will need to open a terminal to run the commands below.

Navigate to your project folder in File Explorer. Click on the address bar, type cmd, and press Enter. This will open a command prompt in that directory. You can also use:

cd [path to your project folder]
# For example:
cd C:\Dev\Discord\MyBot

Create a virtual environment:

Create a virtual environment named venv:

python -m venv venv

Activate the virtual environment:

venv\Scripts\activate

Install the required packages:

pip install -r requirements.txt

Once you've done it once, you'll only need to activate the virtual environment in the future. You'll know it's activated when you see (venv) in your terminal.

Configuration

Create a .env file in the root directory and add your Discord bot token:

DISCORD_BOT_TOKEN=your_token_here

Running the Bot

Run the bot using:

python index.py

Common Issues

Issue	Solution
Python not recognized	Ensure Python is installed and added to your system PATH. You can verify by running python --version in your terminal.
No such file or directory	Make sure you're in the correct directory where index.py is located. Use cd [path to your project folder] to navigate there.
Intents not enabled	Go to the Discord Developer Portal, select your bot, navigate to the "Bot" section, and enable the necessary intents (e.g., "Message Content Intent").
Token not working	Double-check that you've copied the token correctly from the Discord Developer Portal and that it's correctly set in your .env file.
Bot not responding	Ensure the bot is online in your server and that it has the necessary permissions to read and send messages.
Model is too large	Try using a smaller model like TinyLlama 1.1B.
Llama-cpp fails to install	Make sure you have a C++ build environment set up. On Windows, you can install the Build Tools for Visual Studio.

Changing the Model

To change the model, modify the MODEL_PATH variable in index.py to point to your desired model. For example:

llama_model_path = hf_hub_download(
    repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
    filename="mistral-7b-instruct-v0.2.Q4_K_M.gguf"
)

Different models may have different capabilities and performance characteristics, so choose one that fits your needs. Basically anything in GGUF format should work, some might require logging in to Hugging Face. Models are mainly limited by your GPU VRAM.

Example models:

Model	Description
Mistral 7B Instruct v0.2 GGUF	A basic model able to run on basic laptops.
TinyLlama 1.1B GGUF	Really small weight model for quick responses. Doesn't give good responses.