Google launches Antigravity 2.0 with an updated desktop app and CLI tool

· Source: TechCrunch · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

Google has released Antigravity 2.0, an updated version of its agentic coding application, featuring a new desktop app, a CLI tool, and an SDK. The desktop app enables users to orchestrate multiple agents, execute tasks simultaneously, design custom subagent workflows, and schedule background tasks, with integrations for Google AI Studio, Android, and Firebase. Powered by the new Gemini 3.5 Flash model, Antigravity 2.0 also includes native voice command support and a CLI tool for terminal-based agent creation. The Antigravity SDK allows developers to build custom agents, and Google Cloud customers can connect to Antigravity for projects, with custom agent templates available in AI Studio. An export tool in AI Studio facilitates local project continuation. Google is also integrating Antigravity's coding capabilities into consumer products like Search for real-time UI generation and mini-app creation. Additionally, Google is introducing a new AI Ultra plan at $100, offering 5x higher AI limits than the Pro plan, and reducing the price of its top AI Ultra plan from $250 to $200, providing 20x higher limits.

Key takeaway

For developers and engineering teams building agentic coding solutions, Google Antigravity 2.0 offers expanded capabilities through its desktop app, CLI, and SDK. You should explore integrating the Antigravity SDK for custom agent development or utilizing the new desktop app for orchestrating complex workflows, especially if you are already within the Google Cloud ecosystem. Consider the new AI Ultra plans to match your project's AI usage requirements.

Key insights

Google's Antigravity 2.0 expands agentic coding with desktop, CLI, SDK, and Gemini 3.5 Flash integration.

Principles

Method

Antigravity 2.0 enables custom agent development via an SDK, orchestration through a desktop app, and command-line interaction via a CLI tool, all powered by Gemini 3.5 Flash.

In practice

Topics

Best for: Machine Learning Engineer, CTO, VP of Engineering/Data, AI Engineer, Software Engineer, Director of AI/ML

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by TechCrunch.