Free
$0
60 req/min, 1,000 req/day with personal Google account. Includes Gemini 3 models with 1M token context.
Gemini CLI
Google's open-source AI agent for the terminal — powered by Gemini 2.5 Pro.
Gemini CLI is Google's open-source agentic terminal agent powered by Gemini 2.5 Pro. It brings a full AI agent directly into the shell: reads and edits local files, runs shell commands, searches the web via Google Search grounding, and supports MCP. Free for personal use with 1,000 requests/day. ⚠️ Deprecated May 2026 — EOS June 18, 2026. Successor: Antigravity CLI.
+ Pros
- Free for personal use with 1,000 req/day — no credit card required
- 1M-token context window — can ingest an entire codebase in one shot
- Google Search grounding — web-aware responses without a separate tool
- Apache 2.0 open source — 60K+ GitHub stars, forkable and self-hostable
- MCP-compatible — extensible with external tools and APIs
− Cons
- ⚠️ Deprecated: stops working June 18, 2026 — do not invest for new projects
- Successor Antigravity CLI not yet at full feature parity at time of writing
- No persistent memory across sessions by default
- Less autonomous than Cursor, Claude Code, or Devin on complex multi-step tasks
Pricing
Code Assist License
Paid
Higher rate limits via Google Cloud Project with Code Assist license
Vertex AI
Usage-based
Enterprise pricing via Google Cloud Vertex AI
Gemini CLI is Google's open-source AI agent for the terminal, launched at Google I/O on June 24, 2025. It embeds Gemini 2.5 Pro directly into the shell: the agent reads and edits local files, executes shell commands, browses the web with Google Search grounding, and connects to external tools via MCP. The standout feature was a 1-million-token context window — large enough to hold an entire mid-size codebase — offered free for personal use via the Gemini API free tier (1,000 requests/day).
⚠️ On May 19, 2026, Google announced Gemini CLI will stop working on June 18, 2026, replaced by Antigravity CLI — the terminal layer of the Antigravity agentic development platform. The GitHub repository (60K+ stars) will remain public. Existing users are advised to migrate to Antigravity CLI, which ships the same Gemini models with a richer parallel-agent architecture.
Key Features
• ⚠️ Deprecated: EOS June 18, 2026 — successor is Antigravity CLI
• Gemini 2.5 Pro in terminal with 1M-token context window
• Google Search grounding for web-aware responses
• Local file editing and shell command execution
• MCP support: connect external tools and APIs
• Free tier: 1,000 requests/day with personal Google account
• GitHub Actions integration for CI/CD agent workflows
• Apache 2.0 open source · 60K+ GitHub stars
• Cross-platform: macOS, Windows, Linux
Version History
Merged Auto modes into single Auto mode; first-wins agent registration; new agent-tui/tui-tester skills; context files append instead of replace; sandbox and MCP improvements.
Surgical Code Edits via edit tool; session export and import; adaptive token calculator for accurate content size estimation.
Auto Memory Inbox with canonical-patch contract; Gemma 4 models enabled by default; voice mode wave animations; privacy/compliance UX warnings.
Real-time voice mode with cloud and local backends; secure .env loading with workspace trust enforcement; enhanced shell command validation.
Bundled ripgrep for offline search; GitHub-style colorblind themes; MCP resource tools; prompt-driven four-tier memory management.
Signature Snippet
In the terminal: `gemini` then type 'Read all TypeScript files in this project, find all TODO comments, group them by priority, and write a TODOS.md.' Gemini CLI reads the files, extracts TODOs, reasons about priority, and writes the markdown — all in one turn.More in this Space
Vix
Vix is a Go-native, open-source (AGPL-3.0) AI coding agent that slashes token costs by 40-50% using a stem agent architecture and Tree-sitter virtual filesystem. It rethinks the plan/execute loop — keeping LLM cache warm across Explore, Plan, and Execute phases — while shipping Programmable Workflows, Whiteboard Mode with voice AI, MCP server support, and a self-evolving agent that writes its own scheduled jobs and watchers.
Late — High-Leverage AI Agent Orchestration
Orchestrate an entire AI dev team on 5GB VRAM using ephemeral subagents, exact-match diffs, and a zero-dependency Go binary. Works with any OpenAI-compatible model — local or cloud.
Paca
Paca is a free, open-source, self-hosted Scrum board where AI agents work as equal teammates — assigned to sprints, picking up tasks, and collaborating on BDD specs alongside humans. Built as an alternative to Jira and Linear, it treats AI agents as first-class Scrum members.