Installation AutoGPT on Mac M1

All other tutorials did not work for me soo I started at the beginning and wrote my own. Therefore this article is intended for people, who want to try out AutoGPT with a OpenAI API Key.

Step by Step Tutorial

Here you will find a step by step tutorial how to use and install Auto-GPT

1.Step get source code for AutoGPT

Clone The AutoGPT Repository: https://github.com/Significant-Gravitas/Auto-GPT . You con download the zip manually or just do it over the command line via:

$ git clone https://github.com/Significant-Gravitas/Auto-GPT.git

2.Step Create a virtual environment

We need a virtual environment because of of the dependencies for python. Therefore we are using anaconda. To install it on Mac follow https://docs.anaconda.com/free/anaconda/install/mac-os/ . After the installation just do:

$ conda create -n env99 python=3.11 

The installation may ask you about confirming some stuff just agree.

3.Step activate virtual environment

You can now activate the virtual environment with:

$ conda activate env99

4.Step install AutoGPT:

Now after the activation of the virtual environment you are ready to install the requirements. Therefore use the following:

$ cd Auto-GPT/
$ pip install -r requirements.txt

5.Step the Problem

The problem is during the installation I run into this issue here:

$ … ERROR: Failed building wheel for lxml

Or

$ …error: could not build wheels for lxml, which is required to install pyproject.toml-based projects

Or

$ …error: legacy-install-failure

Or

$ … clang-14: error: linker command failed with exit code 1 (use -v to see invocation)
 error: command ‘/usr/local/opt/llvm/bin/clang’ failed with exit code 1

Here is a screenshot:

5.1 Step Solution:

What did the trick for me just run the following:

$ pip install https://github.com/lxml/lxml/releases/download/lxml-4.9.2/lxml-4.9.2-cp311-cp311-macosx_10_15_universal2.whl

6.Step Restarting the Installation

Now that we fixed the lxml dependency we can just rerun the installation command:

$ pip install -r requirements.txt

If it completely successfully it should look something like this:

7. Step get the OpenAI API Key

You can get the Api key here: https://platform.openai.com/account/api-keys
Just click on the button “+ new secret key”.

Give the Api key a name:

Copy the API key:

Save the api key somewhere we need it later.

8. Step add API Key to AutoGPT

Open the the folder Auto-GPT in der Finder. Copy the .env.template file. And Rename the copy of the file to “.env”. If you want to do that via cli just do the following:

$ cd ./Auto-GPT

Of cause if you have already been in your Auto-GPT folder you can skip the above step and only execute this one here:

$ cp ./.env.template ./.env

Now we can go on and edit the just created .env file with vim or sudo. Therefore use the following:

$ nano ./.env

Find the position where it says the “OPENAI_API_KEY=“ and modify it to:

OPENAI_API_KEY=sk-hisfe942389h432h59834ht8439th42389thfweo

IMPORTANT: Here you need to add your api key you created before.

Save the file via keyboard [CONTROL] + O and afterwards press ENTER and exit nano via [CONTROL] + X

9. Step Start Auto-GPT

When you made it until here you are ready to lunch Auto-GPT. Therefore just use the following command to start Auto-GPT:

$ python -m autogpt

Afterwards you should see something like the following screenshot.

In the end after you answered basic questions it will ask you to give it a task:

An example would be “Research the most sold smartphones in 2019”.

And it would start searching the internet about the information. And completed the task writing into a text file a list of the most soled smartphones in 2019.

Conclusion:

I hope you liked the article and it help you. Happy coding.