AutoGPT tutorial: how to set up your own AI-bot in under 30 minutes

Wednesday, April 26, 2023 by Marek

What is AutoGPT

A developer named Significant Gravitas on 30/3/2023 posted on GitHub an open-source Python application, called AutoGPT. It is using GPT-4 as its basis and allows it to act “autonomously” without the need for the user to prompt every action.

AutoGPT has internet access, long-term and short-term memory management, GPT-4 for text generation and file storage, and summarization with GPT-3.5, according to the GitHub post. Anything you can ask ChatGPT, like debugging code and writing an email, you can ask AutoGPT. However, you can ask AutoGPT to complete even more advanced tasks, with fewer prompts and it will first make a plan of achieving it and then will execute it.

AutoGPT can be used to accomplish a variety of tasks, such as research, coding, and creative writing, by assigning roles and goals and using its capability. It is accessed through a web-based interface, which allows users to view data and reports generated by the program.

We encourage you to know more about its capabilities and explore AutoGPT tech page and in-depth AutoGPT guide.

But today we will learn how to set up our own private AutoGPT, without any required coding skill in less than 30 minutes! So, dive in!

Let’s start!

Install Python and visual studio code

Go to python website and download the version for your OS. And install it.

Go to visual studio code and download the version for your OS. And install it.

Get the AutoGPT code

Go to GitHub for the code and download it as a zip file (press the green button). Then extract it, open Visual Studio Code, and open the extracted folder with Visual Studio Code.

Set up an OpenAI API

Get your OpenAI API keys from here.

Go to .env.template and insert your OpenAI API key in the destined field.

Rename .env.template

Press F2 or right-click on the .env.template and rename it to .env

And we’re ready to go!

Let’s test it!

Go to the “terminal” tab in Visual Studio Code and let’s install our dependencies

pip install -r requirements.txt

Now we can run script.

python3 scripst/main.py

And we’re good to go!

Name your AutoGPT, provide a goal and enjoy your private assistant!

AutoGPT and You

As I wrote previously, the potential of AutoGPT exceeds the potential of GPT-4 or ChatGPT. Only the learning ability and access to the internet make it a powerful tool to use it in your everyday workflow.

If you want to learn more from lablab.ai’s AutoGPT tutorial, feel free to. We want you to grow and learn more about AI, as it is a tool to improve our work.

And if you want to build your own new application using AutoGPT API we have an upcoming AutoGPT Hackathon where you can during 24 hours build a working prototype alongside with like-minded people from all around the world.