Neura Market
Open Interpreter logo

Open Interpreter

Free

OpenAI's Code Interpreter in your terminal, running locally.

FreeFree tier
Type
Open Source
Visit Website GitHub (40 stars)

About Open Interpreter

Open Interpreter is an open-source coding agent optimized for low-cost AI models. It runs locally in the terminal, allowing users to execute code, interact with files, and test applications using natural language commands. The tool emulates the agent harness from OpenAI's Codex, with support for multiple harnesses (native, Claude Code, Z-Code, Kimi-Code, etc.) and provider switching. It includes sandboxed command execution on macOS, Linux, and Windows, and a QA skill for operating and testing web and native applications via agent-browser and trycua. Open Interpreter is ACP and Codex compatible, enabling integration with editors and existing Codex workflows. It is free and built to maximize performance with low-cost models like Kimi K3.

Key Features

Emulates agent harness optimized for low-cost models
Supports multiple harnesses: native, claude-code, zcode, kimi-code, swe-agent, and more
ACP compatible; works with ACP-compatible editors and clients
Codex compatible; one-line override for OpenAI Codex SDK
Runs commands inside native sandboxing on macOS, Linux, and Windows
Switch providers and models from the TUI with /model
Inspect and switch harnesses with /harness
Built-in QA skill for testing web apps (agent-browser) and native apps (trycua)
Supports Kimi K3 model with optimized Rust harness

Pros & Cons

Pros
  • Optimized for low-cost AI models, reducing compute expenses
  • Open source (MIT license) and completely free to use
  • Supports a wide variety of agent harnesses and model providers
  • Sandboxed execution for safety and isolation
  • Compatible with ACP and OpenAI Codex, allowing easy integration with existing tools
  • Includes built-in QA capability for automated interface testing
Cons
  • Requires terminal setup and familiarity with command-line interfaces
  • Documentation for advanced features may be limited or community-driven
  • Performance depends on the external model provider and network latency
  • Not a graphical application; no built-in UI beyond the terminal interface

Best For

Running code locally via natural language commands in the terminalTesting and debugging web and native applications using AI agentsBuilding and automating software development workflows with low-cost modelsExperimenting with multiple LLM providers and agent harnesses

FAQ

What is Open Interpreter?
Open Interpreter is an open-source coding agent that runs locally in your terminal. It is a fork of OpenAI's Codex, optimized for low-cost models, and supports multiple harnesses and providers for executing code and testing applications.
How do I install Open Interpreter?
On macOS and Linux, run 'curl -fsSL https://www.openinterpreter.com/install | sh'. On Windows, run 'irm https://www.openinterpreter.com/install.ps1 | iex'. Then type 'i' or 'interpreter' in your terminal to start a session.
What harnesses does Open Interpreter support?
It supports harnesses: native, claude-code, claude-code-bare, zcode, kimi-code, kimi-cli, qwen-code, deepseek-tui, swe-agent, and minimal. Use /harness to switch.
Is Open Interpreter compatible with OpenAI Codex?
Yes. Open Interpreter speaks the same Codex exec protocol. You can override the Codex SDK binary with a one-line change: new Codex({ codexPathOverride: "interpreter" }). It is also ACP compatible.
Can Open Interpreter test applications?
Yes, it ships with a QA skill that lets any model operate and test interfaces. It can drive web apps using agent-browser and native apps using trycua.