#6565th of 87 in Coding agents and assistants
gptme
Terminal-based AI agent that runs shell, Python, browser and file tools
Live demo ↗
(opens in a new tab)Documentation ↗
(opens in a new tab)Website ↗
(opens in a new tab)Repository on GitHub ↗
(opens in a new tab)
- Stars
- 4.4k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Sep 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
gptme is a Python CLI agent that executes shell commands and Python, edits files with patch tools, browses via Playwright and reads images. It works with Anthropic, OpenAI, Google, DeepSeek, OpenRouter, or local models through Ollama and OpenAI-compatible servers. It also ships a Textual TUI, a web UI with REST server, an MCP client and server, ACP support for Zed and JetBrains, and a scaffold for scheduled autonomous agents.
Who it is for: Developers wanting a self-controlled, provider-agnostic terminal coding agent
Strengths
- Provider-agnostic: hosted APIs, OpenRouter, or local Ollama, LM Studio, vLLM and llama.cpp
- Acts as MCP client and server, and as an ACP agent in Zed and JetBrains
- Plugins, skills, hooks and lessons extend behavior without forking
- Runs headless via gptme service init with systemd timers for scheduled agents
Weaknesses
- Executes shell and Python locally; sandboxing (Docker, Wasmtime) is a separate option
- RAG tool needs the separate gptme-rag package
- Large surface area (TUI, web UI, server, desktop app) spread across several repos
- Hosted gptme.ai cloud service exists; its terms and limits are unknown from the README
What it needs
- GPU optional
- Compose
- Needs Playwright (browser tool), Docker or Wasmtime (optional sandboxing), gptme-rag (optional, rag tool)
- Models: Anthropic, OpenAI, Google Gemini, SpaceXAI Grok, DeepSeek
Also in Coding agents and assistants
See all 87| Rank | Project | Score |
|---|---|---|
| 1 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 83 out of 100 |
| 2 | ArchonYAML workflow engine that runs AI coding agents through fixed dev steps | 82 out of 100 |
| 3 | oh-my-openagentTerminal coding agent with multi-model orchestration and git-backed memory | 80 out of 100 |
| #4 | Open SWELangChain coding agent that plans, implements and reviews pull requests | 80 out of 100 |
| #5 | opencodeTerminal coding agent with build and plan modes, plus a desktop app | 79 out of 100 |