Next, you need to download Auto-GPT from GitHub: follow the link for the latest version and download the zip file you find there. When unzipping the archive, duplicate the file named .env.template, and rename it as env. And open it in a text editor. Find the line “OPENAI_API_KEY = your-openai-api-key” and replace “your-openai-api-key” with the actual API key from your OpenAI account.
You can get the API key from here. You only get a few free API calls with ChatGPT, so you’ll want to set up a payment method – you can see the latest prices on this page (at the time of writing it’s a few cents per thousand tokens or bits of information processed). When you set up your payment method, you can also configure payment limits to ensure that AutoGPT doesn’t charge you more than it suits you.
After saving and closing the .env file, you can launch Auto-GPT by right-clicking on its folder and selecting it Open in Terminal. Type “pip install -r requirements.txt” and hit Enters To install the necessary libraries, then when done type “python -m autogpt” and hit Enters to launch Auto-GPT itself. You can then start Auto-GPT anonymously: just tell it what you want it to do, and it will start working.
Its reflection will appear on the screen, and you will be asked to confirm each action in the process Auto-GPT works – if you’d rather agree to multiple actions in a row, type “yx”, where x is the number of actions the bot is happy to perform without your consent (remember all these ChatGPT requests cost you a small amount of money). If you need to terminate a task before it is complete, use Ctrl+C.
If this sounds like a lot of work and you want something easier to tinker with, AgentGPT is a similar project based on Auto-GPT that runs in the browser, no installation required. It’s not quite as advanced as Auto-GPT, but it’s more straightforward for beginners, and gives you an idea of what’s possible with standalone AI agents.
As with Auto-GPT, you need to have an API key for ChatGPT in order for AgentGPT to work properly (and again you will be charged based on usage). You can connect this API key to the tool via settings on the left. Give your agent a name and a goal using the onscreen entry, then choose Deploy Agent to turn it off – tasks appear on the right as they’re created and completed.