A local, open-source implementation of OpenAI's Code Interpreter that lets LLMs run code on your computer through a natural language interface.
Open Interpreter is an open-source tool that enables large language models to run code locally on your computer through a natural language chat interface. It solves the problem of restricted, hosted AI code interpreters by giving users full access to their local environment, internet, and packages without time or file size limits. Essentially, it acts as a local, programmable assistant that can execute tasks like data analysis, file manipulation, and web automation via simple conversation.
Developers, data scientists, and technical users who want to automate computer tasks using natural language without relying on cloud-based, restricted AI services. It's also for those experimenting with local LLMs and seeking a flexible, open-source alternative to proprietary code interpreters.
Developers choose Open Interpreter because it runs entirely locally, offering unrestricted access to system resources, internet, and any package—unlike hosted alternatives. Its open-source nature, model flexibility (support for local and hosted LLMs), and safety-focused interactive approval make it a transparent and powerful tool for AI-driven automation.
A natural language interface for computers
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Open Interpreter runs code directly on your machine, allowing access to all local packages, files, and internet resources without the limitations of cloud-based interpreters like ChatGPT's Code Interpreter.
It connects to various LLMs via LiteLLM, enabling use of both hosted models like GPT-4 and local servers such as LM Studio or Ollama, as shown in the model configuration examples.
By default, it prompts for user confirmation before executing code, with options to configure auto-run or use experimental safe mode, balancing automation with risk mitigation.
Supports a streaming API, custom system messages, and configuration profiles, making it easy to integrate into custom applications like the FastAPI server example provided in the README.
Executing AI-generated code locally can lead to data loss or system damage, and the README admits this with safety warnings and an experimental safe mode that isn't foolproof.
Setting up local LLMs like LM Studio requires additional steps, such as downloading software and configuring servers, which can be challenging for users without prior experience.
LLMs may generate suboptimal or buggy code, and since it runs directly, users need technical expertise to correct issues, reducing reliability for unattended automation.
Open Interpreter is an open-source alternative to the following products: