Google Launches Antigravity CLI for AI-Powered Coding in Terminal Environments
Google described the CLI as a complementary product rather than a replacement for Antigravity 2.0.
Google has launched Antigravity CLI, a lightweight terminal-based interface that brings the company’s AI-powered agentic coding capabilities directly into developer command-line environments.
Google launched Antigravity in November last year, and the newer version earlier this month at Google I/O, its developers conference.
Designed for keyboard-focused developers and remote SSH workflows, the new CLI delivers the same core capabilities available in Antigravity 2.0, including multi-step reasoning, multi-file editing, tool calling and conversation history, but with lower system overhead and faster execution optimised for terminal use.
We’re thrilled to announce the Antigravity CLI, a lightweight way to spin up the same Antigravity agents right from the terminal. 💻
— Google Antigravity (@antigravity) May 25, 2026
It gives you the exact same harness and same models, with a product experience tailored for the command line. It adapts entirely to you: your… pic.twitter.com/0FAYvMZgzN
The company described the CLI as a complementary product rather than a replacement for Antigravity 2.0. While the CLI focuses on speed, efficiency and lightweight workflows, Antigravity 2.0 remains geared toward visual orchestration, project management and broader AI-assisted development tasks.
A key feature of the rollout is the shared agent engine powering both products. According to Google, improvements to reasoning, tool usage and automation capabilities made in one environment will automatically extend across both platforms.
Users will also be able to share preferences and permissions between the CLI and Antigravity 2.0 interfaces.
Google said developers can export conversations from the CLI into Antigravity 2.0 to continue workflows in a more feature-rich interface when needed.
The company is also targeting users of Gemini CLI, offering a migration tool that supports one-time imports of existing extensions, skills and settings to simplify onboarding.
Additional capabilities highlighted by Google include support for slash commands, configurable keybindings, parallel background task delegation and integrations with third-party tools and MCP servers.
