Google Launches Antigravity 2.0 at I/O 2026: A Standalone Agent-First Platform with CLI, SDK, Managed Execution, and Enterprise Support
Google unveiled Antigravity 2.0 at I/O 2026, a standalone desktop application for agent orchestration, marking a significant shift towards multi-agent workflow management. This update includes a CLI, SDK, Managed Agents in the Gemini API, and enterprise support, offering a comprehensive platform for AI-assisted development. The platform integrates with Google AI Studio, Android, and Firebase, providing tools for creating, managing, and deploying AI agents across various environments and catering to diverse developer preferences and enterprise needs.
Google introduced Antigravity 2.0 at I/O 2026, a standalone desktop application designed for orchestrating AI agents. This marks a significant architectural shift, moving Google's developer tooling from IDE-centric assistance to multi-agent workflow management. The platform aims to transform ideas into production-ready applications through an agent-optimized experience.
Antigravity 2.0 features dynamic subagents for parallel workflows, scheduled tasks for background automation, and integrations across Google AI Studio, Android, and Firebase. Scheduled tasks allow agents to operate as persistent automation pipelines, eliminating the need for manual prompting. The platform also includes native voice command support.
Alongside the desktop application, Google released four additional components. The Antigravity CLI offers a terminal-based workflow for creating agents without a graphical interface. The Antigravity SDK provides programmatic access to the agent harness, enabling developers to define custom agent behaviors. The Gemini Enterprise Agent Platform simplifies enterprise workloads, allowing Google Cloud customers to connect Antigravity directly to their projects.
Managed Agents in the Gemini API offer infrastructure-level isolation for agent execution within an isolated Linux environment. These agents, powered by the Antigravity agent harness and built on Gemini 3.5 Flash, support persistent isolated environments for seamless multi-turn sessions and custom agent definitions via markdown files.
The entire Antigravity ecosystem is underpinned by Gemini 3.5 Flash, now the default model, offering improved performance and speed. Google is also expanding developer access with a new Google AI Studio mobile app, an "Export to Antigravity" integration, and native Android support.
Further enhancements include Google Workspace API integration, enabling agents to interact with Workspace services, and direct Google Play Console support within Google AI Studio. Google also introduced a new AI Ultra plan with increased usage limits for Antigravity.
Related articles
Build real agentic apps using CUGA: two dozen working examples on a lightweight harness
CUGA, IBM's open-source Agent Harness, simplifies building agentic applications by handling infrastructure, allowing developers to focus on tools and prompts. It offers pre-assembled components for planning, execution, and state management, significantly reducing development time. CUGA has topped agent benchmarks like AppWorld and WebArena.
OpenAI launches new initiative to help find and patch open source bugs
OpenAI has launched "Patch the Planet," a new initiative in partnership with cybersecurity firm Trail of Bits, to enhance the security of open-source projects. This program aims to assist maintainers in identifying and patching bugs, utilizing OpenAI's AI-powered security tools while reducing the burden on project teams.
PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters
Baidu has released PP-OCRv6, an advanced optical character recognition (OCR) model supporting 50 languages. Available on Hugging Face, this version significantly improves accuracy and efficiency across various parameter sizes, from 1.5 million to 34.5 million, marking a substantial leap in multilingual OCR technology.