G

Gemini CLI

cli Website ↗ 🎁 Start for Free Verified 2026-05-24 · Stale

📊 Scores

Coding Ability
8.5
Value
9.5
Flexibility
8.0
China Access
2.0

🚀 Quick Start

Install Steps

  1. 1 Install Node.js 18+, run: npm install -g @google/gemini-cli
  2. 2 Run: gemini in your project directory (first run opens browser for Google account auth)
  3. 3 After auth, use Gemini 2.5 Pro for free immediately — 1000 requests/day, no credit card needed
  4. 4 Heavy users can get an API Key at aistudio.google.com to bypass free tier limits
Register / Get Account ↗
🇨🇳 China Access Note: Requires Google account; accessing Google services from mainland China requires a proxy; no domestic mirror; China users can try Kimi Code or Trae CN as alternatives

💲 Pricing

Plan Price Limits
Free(个人 Google 账号) $0/mo 60 requests/min, 1000 requests/day; free access to Gemini 2.5 Pro Official →
付费(AI Studio API Key) Billed by Gemini API usage, no daily cap; for heavy users Official →
Vertex AI Enterprise-grade, VPC support, audit logs, regional deployment Official →

Gemini CLI

Free plan available — no credit card required

🎁 Start for Free ↗

✨ Key Features

Code Completion
Chat
Agent Mode
Multi-file Edit
MCP Support
google_search_grounding
web_fetch
shell_exec
plan_mode
sdk_custom_skills
browser_agent
memory_manager
Context Window 100万 token(Gemini 3.1 Pro)
Supported Models Gemini 3.1 Pro (Preview), Gemini 3 Pro, Gemini 3 Flash, Gemini 2.5 Pro, Gemini 2.5 Flash
ide_base 终端
supported_os macOS, Windows, Linux
jetbrains_support
Git Integration

🎯 Best For

Developers who want to experience top-tier models at zero costUsers who need ultra-long context (1M tokens) for large projectsAnyone with a Google account who doesn't want to configure API Keys

📝 Changelog

2026-05-22 v0.43.0 (latest): Surgical code edits — model steered to use edit tool for precise changes; session portability (export/import sessions via CLI flag); adaptive token calculator for accurate content size measurements; multi-agent protocols for local and remote subagent interactions; 80 total changes including OAuth hang fix on headless Linux, parallel tool call streaming fixes, and Git PATH preservation
2026-05-20 v0.43.0-preview.1: Multi-agent architecture preview, ACP-compliant native ToolDisplay properties, sandbox improvements; v0.42.x patch series: OAuth token preservation, snapshot recovery fixes, Sublime Text and Emacs editor integrations
2026-05-05 v0.41.0: Real-time Voice Mode with cloud-based and local processing backends; mandatory workspace trust for headless environments (security); robust shell command validation and core tools allowlist; new ContextManager + AgentChatHistory for improved session handling; auto-memory scratchpad persists across turns for skill extraction
2026-05-01 v0.40.1: Patch fix cherry-picked from v0.40.0
2026-04-28 v0.40.0: Streamlined local Gemma setup via gemini gemma command; prompt-driven memory management replaces legacy MemoryManagerAgent with a more efficient prompt-based editing system

📦 Plans Using This Tool

📖 Related Articles

📚 Related Guides

🔗 Related Tools

💬 Comments

Last Verified: 2026-05-24 Next Review: 2026-06-24 Confidence: high Report Error →