Google Antigravity: A New Agent-Centric IDE for AI-Driven Development

Google has released Google Antigravity, an agentic development platform designed to shift the software creation process from traditional AI-assisted coding to a system where autonomous AI agents lead the development workflow.

Announced on November 18, 2025, alongside the release of Gemini 3, Antigravity is an AI-powered integrated development environment (IDE) that allows developers to delegate complex coding tasks to autonomous agents. The platform is intended for a wide range of users, including professional developers working in large enterprise codebases and hobbyists engaged in vibe-coding.

Technical Architecture and Model Support

Google Antigravity is built as a heavily modified fork of Visual Studio Code. There is ongoing discussion regarding whether the platform is a direct fork of Visual Studio Code or a fork of Windsurf, another AI-oriented editor that itself originates from Visual Studio Code.

The platform is primarily powered by Google’s Gemini 3.1 Pro and Gemini 3 Flash models. However, Antigravity supports a variety of other large language models to provide flexibility in development, including:

• Anthropic’s Claude Sonnet 4.6
• Anthropic’s Claude Opus 4.6
• GPT-OSS-120B, an open-source variant of OpenAI models

The Agent-First Paradigm

The platform introduces what Google describes as an agent-first paradigm. This approach differs from standard AI code assistants by granting agents greater autonomy to operate across the development process rather than simply suggesting lines of code.

Antigravity features two primary interface views. The Editor view provides a standard IDE experience similar to PyCharm or Visual Studio Code but includes an agent sidebar. This interface is comparable to tools such as GitHub Copilot or Cursor, allowing the agent to work alongside the developer in a coordinated manner.

Availability and System Requirements

Google Antigravity was released in public preview on November 18, 2025, with version 1.11.2. A subsequent preview release, version 1.21.9, was issued on March 30, 2026.

The IDE is currently available free of charge during its preview period and includes generous rate limits for the use of Gemini 3.1 Pro. The platform is compatible with the following operating systems:

• 64-bit Windows 10 or later
• macOS Monterey 12 or later
• 64-bit Linux with glibc 2.28 or later and glibcxx 3.4.25 or later

By combining a familiar editor experience with a task-oriented agent interface, Google aims to enable developers to operate at a higher level of abstraction, focusing on task completion rather than manual syntax implementation.