ChatGPT IRC Bot

ChatGPT IRC Bot is a simple IRC bot written in Python. It was initially forked from the knrd1/chatgpt but turned into its own project. It connects to OpenAI endpoints to answer questions or generate images and uses official bindings from OpenAI to interact with the API through the HTTP requests. You can find more details about API here.

Prerequisites

1. Create an account and obtain your API key: https://platform.openai.com/account/api-keys
2. Install Python3 and the official Python bindings (openai: 0.28.0, 0.28.1; pyshorteners)

   Note that only version 0.28.x of openai is supported, and the latest version is 0.28.1.

   • Debian/Ubuntu

     apt install python3 python3-pip
     pip3 install openai==0.28.1 pyshorteners
     

   • RedHat/CentOS

     yum install python3 python3-pip
     pip3 install openai==0.28.1 pyshorteners
     

   • FreeBSD

     pkg install python311 py311-pip
     pip install openai==0.28.1 pyshorteners
     

Installation

Clone the package using the below command. It will copy files into the ChatGPT-IRC-Bot directory, which you can later rename.

git clone https://github.com/oiramNet/ChatGPT-IRC-Bot.git

Configuration

ChatGPT IRC Bot uses a plaintext file as its configuration file. The package includes an example configuration file (chat.conf.sample) set to connect to IRCnet. You can copy and modify it.

cd ChatGPT-IRC-Bot
cp chat.conf.sample chat.conf

Variable context is optional, you can leave it blank or enter what you want the bot to know and how you want it to behave. This will only work with the models connecting to the endpoint /v1/chat/completions (see below).

[openai]
api_key = sk-XXXXXXXXXXXXXXX

[chatcompletion]
model = gpt-4o-mini
role = user
context = 
temperature = 0.8
max_tokens = 1000
top_p = 1
frequency_penalty = 0
presence_penalty = 0
request_timeout = 60

[irc]
server = open.ircnet.net
port = 6667
ssl = false
channels = #oiram
nickname = SampleBot
ident = samplebot
realname = Sample Bot
password = 

Model endpoint compatibility

ChatGPT IRC Bot can use one of three API models:

• Models that support endpoint /v1/completions (Legacy)

  gpt-3.5-turbo-instruct, babbage-002, davinci-002

• Models that support endpoint /v1/chat/completions

  gpt-4o-mini, gpt-4o, gpt-4, gpt-4-turbo, gpt-4-turbo-preview, gpt-3.5-turbo

• Models that support the creation of an image using endpoint /v1/images/generations

  dall-e-2, dall-e-3

We suggest starting experimenting with gpt-4o-mini model. More details about models can be found here.

Running bot

To start the bot, you can just run the command below.

• Debian/Ubuntu/RedHat/CentOS

  python3 chatgpt.py chat.conf
  

• FreeBSD

  python3.11 chatgpt.py chat.conf
  

You can use the screen command to run it in the background and keep it running even after you log out of your session.

• Debian/Ubuntu/RedHat/CentOS

  screen python3 chatgpt.py chat.conf
  

• FreeBSD

  screen python3.11 chatgpt.py chat.conf
  

To detach from the screen session (leaving your ChatGPT IRC Bot running in the background), press Ctrl + A followed by d (for "detach"). If you need to reattach to the screen session later, use the following command:

screen -r

Interaction

ChatGPT IRC Bot will interact only if you direct your question using its nickname.

12:34:12 < user> SampleBot: how are you?
12:34:13 < SampleBot> I'm just a computer program, so I don't have feelings, but I'm here and ready to help you! How can I assist you today?
12:56:21 < user> SampleBot: do you like IRC?
12:56:22 < SampleBot> As an AI, I don't have personal preferences or feelings, but I can provide information about IRC! Internet Relay Chat (IRC) is a popular protocol for real-time communication and has been used for
decades. Many people appreciate it for its simplicity and the sense of community it fosters. If you have specific questions or topics about IRC, feel free to ask!

If you set the model to dall-e-2 or dall-e-3, the ChatGPT IRC Bot will return a shortened URL to the generated image.

13:14:05 < user> SampleBot: red apples on the table
13:14:35 < SampleBot> https://tinyurl.com/1a2b3c4d

Docker

To build the Docker image, you can use the following command:

docker build -t my-chatgpt-app .

To run the Docker container, you can use the following command:

docker run -it my-chatgpt-app

To detach from a running Docker, press Ctrl + P. While holding down Ctrl, press Q. To reattach to the container later, use the following command:

docker attach <container_id>