This content is not affiliated with, endorsed, sponsored, or specifically approved by OpenAI and OpenAI is not responsible for it.

I like the web interface when I'm using chatGPT on my computer. On the other hand I never use it on mobile (it's to much work going to web browser, sign-in and chat). So I decide to write a very small Telegram Bot (< 80 LOC) using OpenAI API and aiogram.

OpenAI offers some free credit for their API ($18.00 in credits for 3 month). This bot use gpt-3.5-turbo endpoint for chat completions which means that requests cost $0.002 per 1k tokens (see blog post). So you can deploy this bot on your machine and enjoy for three month your personal instance of chatGPT.

1. Clone this repo and go in it: git clone https://github.com/S1M0N38/Telegram-GPT.git && Telegram-GPT
2. Install dependencies: python -m pip install -r requirements.txt
3. Create new Telegram Bot obtaining a token (e.g. 2837489710:GJjkajsJFmmjkjDJKSekj00JKKjkjahdja2)
4. Create new OpenAI API key (e.g. sk-askldj9382kjdklqLlkaj92lLanEREybjae36HSMAnd2iudj)
5. Create new file .env with the following env variables (use token and api-key obtain from the previous steps)

export TELEGRAM_TOKEN="2837489710:GJjkajsJFmmjkjDJKSekj00JKKjkjahdja2"
export OPENAI_TOKEN="sk-askldj9382kjdklqLlkaj92lLanEREybjae36HSMAnd2iudj"

5. Load .env and start the bot: source .env && python bot.py

Start a conversation with your bot. You can start a new conversation with /start.