Open Interpreter: An interesting AI tool that runs a code interpreter like ChatGPT locally

On the heels of Auto-GPT and the Code Interpreter API, a new open source project is making waves in the AI ​​community. The project is called Open Interpreter and was developed by Killian Lucas and a team of open source contributors. Combine ChatGPT plugin functionality, code interpreter, Windows Copilot, and more to make AI a ubiquitous solution on any platform. You can do anything you can think of with Open Interpreter. You can interact with the system, files, folders, programs, the Internet, and basically everything at the OS level directly from the friendly terminal interface. If you’re interested, learn how to set up and use Open Interpreter locally on your PC.

Things to remember before proceeding 1. To take full advantage of Open Interpreter, you must have access to a GPT-4 API key. GPT-3.5 model does not work at all and throws multiple errors while running the code. Yes, running the GPT-4 API is expensive, but it opens up many new utilities on your system. We recommend that you do not run your model locally unless you are familiar with the build process. This project currently has bugs, especially in the local model on Windows (at least for us). Also, larger models require stronger hardware specifications for better performance.

Set up the Python environment

1. First, you need to install Python and Pip on your PC, Mac, or Linux computer. Follow the linked guide for detailed instructions.

2. Next, open Terminal or CMD and run the command below to update Pip to the latest version .

 python -m pip install -U pip

3. Next, install Open Interpreter on your machine by running the following command:

 pip install open-interpreter 

Set up Open Interpreter on your PC

1. Once Open Interpreter is installed, run one of the following commands depending on your preference :

• For GPT-4 models . OpenAI must have access to the GPT-4 API.

 interpreter

• For GPT-3.5 model . Available to free users.

 interpreter --fast

• Run the Code-llama model locally. Available for free. Computers require good resources.

 interpreter --local

2. We’ll use the OpenAI GPT-4 model here, but you can choose GPT-3.5 if you don’t have access to the API. Next, obtain your API key from the OpenAI website . Click “Create new private key” and copy the key.

3. Paste the API key into Terminal and press Enter. By the way, you can exit Open Interpreter at any time by pressing “Ctrl + C”.

How to use Open Interpreter on your PC

1. To start using Open Interpreter, I asked to set my system to dark mode and it worked fine. Since I’m using Windows, a registry key was created and the theme changed seamlessly.

2. Next, I asked them to create a simple web-based timer app . And in no time, the app was ready. This can be a great way to make money by harnessing the power of AI.

3. We then asked the open interpreter to summarize the local text document . This worked fine.

4. I asked them to convert the PDF file to DOCX and it worked fine.

5. Next, we quickly pulled data from the Internet and displayed the stock prices of Apple and Microsoft in a visual chart. Open Interpreter allows you to experiment with many use cases like this, which can be very interesting.

Nifty tips for using Open Interpreter locally

1. Open Interpreter always asks for permission before running code. Beyond a certain point, it can become frustrating. If you want to avoid this, start Open Interpreter as follows.

 interpreter -y

2. Next, you can permanently set the OpenAI API key in the command line interface as follows: This saves a lot of time. Replace ” your_api_key ” with your actual key.

• windows

 setx OPENAI_API_KEY your_api_key

• macOS and Linux

 export OPENAI_API_KEY=your_api_key

3.If you want to use a different model locally , you can define it like this: The model must be hosted on Hugging Face and the repository ID must be listed below. You can find the best open source AI models from the linked article.

 interpreter --model tiiuae/falcon-180B

If you think your project is too complex or buggy, check out our list of the best AI coding tools. These tools give you the best coding experience, whether it’s your favorite IDE or a simple code editor.