What is AutoGPT, Automated version of ChatGPT; Everything you need to know

Adithya Thatipalli
3 min readApr 15


Release of AutoGPT stormed across the internet, but what is this AutoGPT.

What Is Auto-GPT?

AutoGPT is an experimental open-source project that enables you to specify a certain job and a number of goals and then let GPT-4 automatically construct and carry out all the tasks required to meet your goals like (E.g., “ Viral Video Niche”) and a bunch of goals (e.g., “research the most successful Instagram trends”, “create a calendar”, “save the summary to a file,” etc.) and thats it.

Let GPT-4 automatically create and perform all the necessary tasks that are needed to achieve your goals. It is capable of the following:

  • Searching the web for research
  • Scraping data
  • Exporting files to a database
  • spawning new instances of GPT-3.5 bots with special roles to accomplish sub-tasks (!)

AutoGPT is a research project, so it has errors and shortcomings, particularly when it encounters unfamiliar roles and objective descriptions, but it’s already a fantastic demonstration of GPT-4’s capabilities and the possibility of a modular method to combine numerous AI agents.

How To Setup Auto-GPT


a) A system with Python 3.8 or later

b) OpenAI API keys (if you have an account with OpenAI, you’ll find them here; If you don’t have an OpenAI account, you can follow these instructions: https://beta.openai.com/docs/api-reference/authentication1

1. Clone the Git Hub Repo

If you have GIT installed on your computer, simply run this in your terminal:

  • Clone the repository from Github or download the ZIP file of the source code.
git clone https://github.com/Torantulino/Auto-GPT.git

Alternatively, you can download the ZIP file of the source code here. In the top right corner, click Code and then Download Zip.

2. Install the required libraries

Navigate to the directory where your copy of Auto-GPT resides (it’s called “Auto-GPT”) and run it.

pip install -r requirements.txt

3. Add API Keys

Rename the below file from




and fill in your OPENAI_API_KEY.

Alternatively, you can use PineCone API keys

4. Run Auto-GPT

Now you can run your AutoGPT

python scripts/main.py

But, if you do not have access to GPT-4 or don’t want to use it, run the following command instead.

python scripts/main.py --gpt3only
  • Decide whether you want to continue with the last AI agent you created or create a new one.
  • If you create a new one, give it a name and describe its role.
  • Give it a bunch of goals (up to five) that you want it to achieve.
  • Authorize its next steps by typing “y” or “n” when prompted. You can also type “y -N” to let it run multiple commands without asking for permission.

5. Review the outputResults

To check Auto-GPT’s results, you can either:

  • Have a look at its log files in the folder “/outputs/logs”.
  • Check the files it created in “/auto_gpt_workspace”.

Final Thoughts

AutoGPT is an open-source Python programme that employs the GPT-4 language paradigm to build self-prompting AI bots that can carry out a variety of online tasks.

By giving GPT-based models memory and autonomy, AutoGPT is revolutionizing the field of artificial intelligence by enabling them to complete tasks on their own and gain knowledge from their mistakes.

Numerous sectors could undergo a revolution thanks to the use of AutoGPT agents, which would greatly improve production, accuracy, and ultimately the top and bottom.

This experiment aims to showcase the potential of GPT-4 but comes with some limitations:

  1. Not a polished application or product, just an experiment
  2. May not perform well in complex, real-world business scenarios. In fact, if it actually does, please share your results!
  3. Quite expensive to run, so set and monitor your API key limits with OpenAI!



Adithya Thatipalli

Security Engineer by Day, Cloud and Blockchain Learner during Night