Gemini CLI
Gemini CLI is an open-source command-line tool that brings Google’s Gemini models into your terminal. It supports code generation, editing, project scaffolding, multimodal input, tool execution, and slash-command workflows. You can use it to summarize code changes, query large codebases, refactor code, automate rebases, or even create apps from PDFs and sketches.
Key features include:
- Custom slash commands via TOML files
- MCP integration with
/mcpcommands and tool discovery - OAuth, Vertex, and API key authentication
- Shell passthrough (!), file injection (@), and vim-mode editing
- Full local config and memory override support via GEMINI.md files
Gemini CLI is highly extensible, supports CI/CD workflows with headless auth, and includes built-in privacy, telemetry, and version controls.
No discussions yet
Be the first to start a discussion about Gemini CLI
Developer
Google’s AI division building multimodal models and tools like Gemini, Gemini Pro, Imagen, and Veo.
Pricing and Plans
(Free)
Free
Free
- Unlimited local usage with personal Google account
- 60 requests/min, 1,000/day (Gemini API free tier)
- 100/day (Gemini Pro 2.5 API free tier)
- Vertex AI express mode available without billing
System Requirements
Operating System
LINUX, MACOS, WINDOWS
Memory (RAM)
4GB minimum
Processor
x86 or ARM CPU
Disk Space
200MB
AI Capabilities
Multimodal generation
Slash command execution
Large context code analysis
Shell command passthrough
Custom prompt injection
Model memory management
Tool-based file edits