openinterpreter /openinterpreter
A coding agent for open models like Kimi K3 and GLM 5.3
★ 68,889⑂ 5,947👁 452⑃ 783● 1,131▤ 78
+3
日增长
+46
周增长
68,889
stars
5,947
forks
452
watchers
783
prs
1,131
issues
78
releases
README
Open Interpreter
A coding agent optimized for low-cost models. Blog post ↗
English • Español • 简体中文 • 日本語
[!NOTE]
Today: Kimi K3 is here. We have reimplemented the provider-recommended
Kimi Code harness in Rust, giving you
maximum K3 performance with a Codex-like interface.
Kimi Docs →
Installation
macOS and Linux:
bash
curl -fsSL https://www.openinterpreter.com/install | sh
Windows:
powershell
irm https://www.openinterpreter.com/install.ps1 | iex
Then type i or interpreter in your terminal to start a session.
Harness Emulation
Open Interpreter is a fork of OpenAI's Codex, with a focus on emulating the agent harness that gets the best performance out of low-cost models.
Use /harness to switch the active harness:
text
/harness
native
claude-code
claude-code-bare
zcode
kimi-code
kimi-cli
qwen-code
deepseek-tui
swe-agent
minimal
Read more in the harness docs and provider setup guides.
ACP compatible, Codex compatible
Open Interpreter works in ACP-compatible editors and clients. Configure the client to launch interpreter acp; see the ACP guide for examples.
Already building with OpenAI's Codex SDK? Keep the SDK and make a one-line
binary override:
diff
-const codex = new Codex();
+const codex = new Codex({ codexPathOverride: "interpreter" });
Open Interpreter speaks the same Codex exec protocol. See the SDK guide and run scripts/test-codex-sdk-compat.sh for a local, provider-free compatibility check.
Want a small browser UI over raw Open Interpreter? Start with the
local streaming web chat guide. For a complete
desktop, browser, or headless client, use
Interpreter Workstation.
Portable by default
Open Interpreter should fit into your existing agent setup instead of trapping
it in an Open Interpreter-only format. The product goal is to prefer shared,
tool-neutral standards and directories, keep user-authored data in readable
files, and make moving to or from another compatible agent straightforward.
Today that includes repository AGENTS.md, shared .agents/skills directories,
MCP, ACP, and the Codex exec protocol. Product-specific storage under
~/.openinterpreter is reserved for configuration and runtime state that does
not yet have a practical shared standard. Legacy product-specific skill
directories remain readable for compatibility, but new skills belong in
.agents/skills or ~/.agents/skills.
See the portability guide for the current boundary and
the rules for evolving it.
Computer Use
Open Interpreter ships with a QA skill that lets any model operate and test interfaces. It can drive web apps in a real browser with agent-browser, or operate and test native apps with trycua.
Features
- Runs commands inside native sandboxing on macOS, Linux, and Windows.
- Switches providers and models from the TUI with /model.
- Runs any selected OpenAI-compatible provider through Chat Completions with
interpreter --chat-completions or interpreter exec --chat-completions.
- Inspects or sTop issues
#1908DeepSeek Chat Completions profile fails after Allow Always/tool approval5 comments · 2026-09-14#1947Windows 11: Interpreter 0.2.199 repeatedly spawns blank windows; EPERM during skill initialization3 comments · 2026-10-19#1928Does not work with Gemini due to hard-coded 'store' key3 comments · 2026-09-29#1962Provider catalog generator ignores explicit API base URL overrides2 comments · 2026-11-01#1956feat(security): Sub-35us AST Invariant Gate to prevent catastrophic shell calls & secret leaks2 comments · 2026-10-29#1949Chat-wire providers show the OpenAI "high-risk cyber activity" warning whenever the routed model differs2 comments · 2026-10-20#1952Duplicate agent response when using ACP server (e.g. via Zed)1 comments · 2026-10-22#1938Desktop app ignores function-calling disablement and loops raw JSON with Ollama1 comments · 2026-10-03
Discussions
#1983Chopper: a small open model that cuts only the needed sentences out of retrieved text0 comments · 2026-11-13#1946Open Interpreter is included in the AgentManager tool catalog0 comments · 2026-10-16#1920grunz: chat + coding agent on abliterated/uncensored open models0 comments · 2026-09-25#1913Explicit Edit Benchmark0 comments · 2026-09-19#1864cowork-to-code-bridge — a file-queue bridge so a sandboxed agent can hand work to a local one (no socket, no open port)0 comments · 2026-08-05#1793Llama.cpp configuration3 comments · 2026-06-29#1792cowork-to-code-bridge — let a sandboxed agent safely reach Open Interpreter's real machine (async, file-based RPC)0 comments · 2026-06-28#1789cowork-to-code-bridge — async bridge from a sandboxed agent to Claude Code on your local machine0 comments · 2026-06-25
Latest release
Open Interpreter 0.0.61 rust-v0.0.61 2026-11-11Open Interpreter 0.0.61 Models - No model-provider catalog or request-admission behavior changes are included in this release. Release - Attest Linux Open Interpreter archives and their checksum manifest in the original tagged build. - Verify that original build provenance before publishing existing artifacts.