Open Interpreter

Open Interpreter is a locally installed terminal coding agent for developers on macOS, Linux and Windows. It's a fork of OpenAI's Codex that focuses on adapting agent behavior to lower-cost models. Commands run on your computer inside native sandboxing; model requests go to your selected provider, so local installation alone doesn't mean local inference.

You can switch models and providers within the terminal interface, including services with an OpenAI-compatible Chat Completions API. The agent also lets you choose model-specific behavior, known as a harness. Its Rust implementation of the provider-recommended Kimi Code harness supports Kimi K3, while provider guides also cover DeepSeek and Z.AI models.

For developers who need the agent to check its work in an interface, a built-in QA skill can operate web apps in a real browser through agent-browser. It can also operate and test native apps through trycua.

Open Interpreter works with editors and clients that support Agent Client Protocol (ACP). Applications built with OpenAI's Codex SDK can use it through the compatible Codex exec protocol. It reuses repository AGENTS.md instructions, shared skills and MCP connections, and supports hooks and permission controls.

The license is Apache 2.0. Product-specific configuration and session state stay in local storage, while hosted model calls use the provider's credentials.