Home > Technology peripherals > AI > With 13k stars, the open source version of the code interpreter tops the GitHub hot list and can be run locally and access the Internet.

With 13k stars, the open source version of the code interpreter tops the GitHub hot list and can be run locally and access the Internet.

WBOY
Release: 2023-09-17 19:13:01
forward
1202 people have browsed it

In July this year, OpenAI opened a powerful plug-in, Code Interpreter, which can generate and execute code solutions based on the user’s natural language needs, and ultimately complete the analysis of data and creation of charts. , editing files and other tasks.

However, OpenAI’s Code Interpreter has some specific limitations. The plugin runs in a controlled remote environment, is hosted, and is closed source, meaning the developer has limited control over local settings. In addition, Code Interpreter does not have access to the Internet and has certain limitations on the size of uploaded files and runtime

Now, an open source project called Open Interpreter has solved these problems. Open Interpreter can run in a local environment with Internet access, has no restrictions on file size or runtime, and can leverage any package or library. The project has only been online for one week and has already received 13k stars

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

  • ## Project address: https://github.com/ KillianLucas/open-interpreter/#commands
  • Project homepage: https://openinterpreter.com/

Using GPT-4 API combined with Open Interpreter, Python code can be executed in the local environment or Google Colab. This makes GPT-4 very powerful. Through Open Interpreter, users can control computers to complete various tasks through natural language, such as summarizing PDFs, visualizing data sets, and controlling browsers

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

## Taking Apple computers as an example, users can enter natural language commands in the Mac terminal interface and open the interpreter to meet the user's needs. For example, you can switch your system to dark mode by entering the command

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

Summary summary pdf long article:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

Convert Word files to PDF files:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

View the to-do list on Calendar software Do things and email them to someone:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

You can even write a simple Pomodoro software:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

In addition, Open Interpreter can solve mathematical problems by executing Python code. For example, to solve the equation 10x 14 = 12:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

In terms of installation and use, the installation of Open Interpreter is very simple, requiring only one line of code:

pip install open-interpreter
Copy after login

After the installation is complete, enter interpreter in Terminal to start using Open Interpreter.

Open Interpreter allows LLM to run code in Python, Javascript, Shell and other languages ​​locally. If there is no GPT-4 API, you can also use the open source model Code Llama through the following command line:

interpreter —local
Copy after login

or use the following command line to use gpt-3.5-turbo:

interpreter —fast
Copy after login

Open Interpreter 迅速在机器学习社区爆火,很多网友都称赞其是一个非常优秀的项目:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

还有网友关心 Open Interpreter 的安全问题。对此,项目作者回应道:「Open Interpreter 在运行任何代码之前都会要求进行验证」。

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

项目作者还表示后续将把 Open Interpreter 开发成一个桌面应用程序:

狂揽13k star,开源版代码解释器登顶GitHub热榜,可本地运行、可访问互联网

对于那些对这个强大又有趣的开源项目感兴趣的读者们,不妨赶快去试试看吧

The above is the detailed content of With 13k stars, the open source version of the code interpreter tops the GitHub hot list and can be run locally and access the Internet.. For more information, please follow other related articles on the PHP Chinese website!

Related labels:
source:51cto.com
Statement of this Website
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
Popular Tutorials
More>
Latest Downloads
More>
Web Effects
Website Source Code
Website Materials
Front End Template