Gemini CLI
Google's open-source AI agent that runs Gemini models in your terminal
Google's open-source AI agent that runs Gemini models in your terminal
Gemini CLI is an open-source AI agent from Google that works in your terminal. You ask in plain language; it reads and edits files, runs shell commands, fetches web pages and uses Google Search. It connects to MCP servers, takes extensions and custom commands, reads project context from GEMINI.md files and runs in scripts without interaction.
It runs on your computer; the Gemini models run in Google's cloud, so it needs the internet. It is free under the Apache 2.0 licence. It needs Node.js 20 or later and runs on macOS 15+, Windows 11 24H2+ and Ubuntu 20.04+. Install with npm, Homebrew or MacPorts.
Google says that since 18 June 2026, Gemini CLI no longer serves unpaid users or Google AI Pro, Ultra and Google One subscribers; they move to Antigravity CLI. It still works with Gemini Code Assist Standard (1,500 requests a day; $19 per user a month on a 12-month term, $22.80 monthly) and Enterprise (2,000 a day; $45 or $54), and with paid Gemini API keys or Vertex AI, billed by use.
Developers who work in the command line. Since June 2026 Google offers it to organizations with a Gemini Code Assist Standard or Enterprise licence and to users of paid Gemini API keys or Vertex AI.
Open source under Apache 2.0, and runs with a paid Gemini API key, billed by use.
Uses Google's Gemini models with Google Search built in, through Gemini Code Assist or Vertex AI.
Discussion
Sign in to join the discussion