Google Antigravity

Google Antigravity is Google's agentic development platform, released in public preview on November 20, 2025. It combines a full-featured AI-powered IDE with a dedicated agent orchestration interface, designed to let developers operate at a higher, task-oriented level rather than line-by-line.

**Two ways to work**: - **Editor View** — A state-of-the-art AI code editor with tab completions, inline commands, and the familiar synchronous workflow developers already know. - **Manager Surface** — An agent-first interface where you spawn, orchestrate, and observe multiple AI agents working asynchronously across different workspaces. Agents can plan, execute, and verify tasks autonomously across editor, terminal, and browser.

**Multi-model flexibility**: Antigravity offers generous access to multiple frontier models within a single IDE: - Gemini 3.1 Pro and Gemini 3 Flash (Google's own) - Claude Sonnet 4.6 and Claude Opus 4.6 (Anthropic) - GPT-OSS models (open-weight)

Different agents within a session can be assigned different models — use Gemini 3 Pro for architecture, Claude Sonnet for implementation.

**Agentic workflows**: Agents autonomously write code, launch apps in the terminal, open the browser to test and verify output, and communicate results via Artifacts (screenshots, walkthroughs) — all without requiring constant human input.

**Free during preview**: Antigravity is completely free for individuals with a personal Google account during the public preview phase. Weekly-based rate limits apply, but access to Gemini 3 Pro, Claude Opus 4.6, and other models at zero cost is genuinely unusual in the market.

Available for macOS, Windows, and Linux.