AI Agent Platforms / All platforms / Open Interpreter
Open Interpreter
Coding agent optimized for low-cost models that emulates multiple agent harnesses.
Open Interpreter is a coding agent optimized for low-cost language models, built as a reimplementation of OpenAI's Codex. It runs commands in native OS sandboxes and lets users switch models and agent harnesses from the terminal.
| Category | Coding agents |
| License | Apache-2.0 |
| Hosting | runs locally |
| Repository | github.com/openinterpreter/open-interpreter |
| Website | openinterpreter.com |
| Last verified | 2026-09-02 |
Documented features
- Runs commands inside native sandboxing on macOS, Linux, and Windows
- Switches providers and models from the TUI with the /model command
- Supports models such as Kimi K3, DeepSeek, Qwen, and Z.AI/GLM
- Emulates multiple agent harnesses (native, claude-code, kimi-code, deepseek-tui, swe-agent, minimal) via /harness
- Tests web and native apps through a built-in QA skill
- Compatible with the Agent Client Protocol (ACP) and the Codex SDK
- Reuses shared AGENTS.md instructions and .agents/skills directories
- Applies permission controls to sandboxed execution
Feature summaries come from the project's own repository and documentation and describe the software as documented on the verification date, not an evaluation.