Gemini API CLI

Gemini API CLI (gemini-api) is Google's experimental command-line tool that lets coding agents such as Claude Code, Codex and Cursor use the Gemini API's multimodal capabilities straight from the shell.

What it is

• Form: zero-dependency Go binary, Apache-2.0 (github.com/google-gemini/gemini-api-cli, v0.4.3)
• Capabilities: generate images/video/music/speech, analyze and transcribe media, run Managed Agents
• Timeline: announced Oct 19; heavy EN/CN/JP discussion on X Oct 23–24

Not to be confused with

It is not Gemini CLI (the interactive coding agent) nor Antigravity CLI; it is an API tool layer for other agents to call.

How to use it

1. Download the binary for your platform from GitHub Releases and add it to PATH
2. Set a Gemini API key (from Google AI Studio)
3. Allow your agent to call gemini-api, e.g. have Claude Code generate an image or transcribe audio

Official intro: aistudio.google.com/learn/introducing-the-gemini-api-cli

Experimental: flags may change, check the README; usage is billed per API call.