Overview

• Gemini CLI enables natural language prompts in the terminal for coding tasks such as writing new features, debugging, and file manipulation, as well as non-coding uses like research report generation and multimedia creation.
• The agent employs the Gemini 2.5 Pro reasoning model with a one-million-token context window, allowing it to understand entire codebases and maintain extensive conversation history.
• Google’s free tier grants developers up to 60 model requests per minute and 1 000 per day, which company data shows is double typical internal usage patterns.
• The open-source release under Apache 2.0 invites community contributions and code auditing, while built-in security prompts let users approve command execution on a case-by-case basis.
• Support for the Model Context Protocol enables custom extensions and integrations with services like Google Search, Veo and Imagen, and Google plans to extend the same capabilities into VS Code via Gemini Code Assist with paid enterprise options available.