原文 · 未翻译
All the news from the Google I/O 2026 Developer keynote
Facebook
Twitter
LinkedIn
Mail
We've transitioned from AI that simply assists you, to agents that can independently navigate complex tasks across your entire workflow. This year at I/O, we announced our Gemini 3.5 series of models and upgraded Antigravity, our agent-first development platform, with new capabilities to orchestrate and build agents. And we showed how we're making it easier than ever for agents to help you build modern and performant Android and web apps with all new tools and skills.
We're bringing the models, infrastructure, and platforms together across Google to help you build anything you can imagine.
Link to Youtube Video (visible only when JS is disabled)
Orchestrating and building powerful agents
Orchestrating agents: With Antigravity 2.0 and the all-new Antigravity CLI, you now have two powerful surfaces for incredible productivity gains. You can spin up specialized subagents to tackle complex workflows, all protected by built-in cross-platform terminal sandboxing, credential masking, and hardened Git policies.
New Google AI Studio integrations: Google AI Studio now includes native Kotlin support to vibe code Android apps. With Google Workspace integrations and a one-click deploy to Cloud Run along with support for Firebase services, you can now build and launch full-stack apps directly within AI Studio. And if you want to continue building, you can also seamlessly export your complete project state to Google Antigravity.
Build managed agents with Gemini models: Managed Agents in the Gemini API removes the friction of infrastructure setup, delivering the power of the Antigravity agent harness via managed agents. A single API call provides a fully provisioned agent with a remote sandbox.
Host your own Antigravity agent: We’re also giving you programmatic control over the Antigravity agent harness with the new Antigravity SDK, so you can fully customize the agent and deploy it on your own infrastructure.
Building high-quality Android apps with agents
Android CLI & skills: The stable Android CLI enables your AI agents to tap directly into the “heavy-lifting” power of Android Studio. It can handle tasks like downloading the Android SDK, running your app on Android devices, and more, so you can create high-quality Android apps using any agent, LLM, and tool of your choice. We also open-sourced Android skills to help LLMs execute best practices for complex workflows and APIs, like migrating to Jetpack Compose, or Jetpack Navigation 3 migration.
Android Bench: You need AI that understands the nuances of Android, which is why we created Android Bench, our LLM leaderboard for Android development tasks. This week, we added open-weight models such as Gemma 4 to the leaderboard, so you can see how other LLMs measure up.
Migration agent: We previewed an exciting new Android Studio feature that migrates your app code to a native Kotlin Android app, regardless of whether your source is React Native, a web framework, or iOS. The agent analyzes code and does the heavy lifting, turning migrations that would have taken weeks into just hours.
Redefining web development in the agentic era
WebMCP: WebMCP is a proposed open web standard that allows developers to expose structured tools, like JavaScript functions and HTML forms, so browser-based AI agents can execute complex tasks with greater speed, reliability, and precision. The experimental WebMCP origin trial starts in Chrome 149, with support for Gemini in Chrome coming soon.
Modern Web Guidance: Modern Web Guidance will help you build more performant, accessible, and secure web experiences by providing your coding agents with a set of expert-vetted skills. Launching in early preview, it supports over 100 use cases. It also integrates directly with Baseline and fallbacks to use within your chosen Baseline target. Install Modern Web Guidance with a single click in Antigravity or via CLI with `npx modern-web-guidance install`.
Chrome DevTools for agents: We’re bringing Chrome DevTools’ capabilities to AI agents with Chrome DevTools for agents, helping you scale your workflow with verifying, debugging, and optimizing code in real time. Your agent can automate quality audits, emulate real-world user experiences, hand over sessions with auto-connect, and more without manual oversight.
HTML-in-Canvas: With the new HTML-in-Canvas API, available in origin trial, developers can build immersive, 3D experiences that remain fully searchable, accessible, and interactable. By integrating real DOM elements directly into a canvas with WebGL and WebGPU, this declarative API brings high-fidelity, app-like performance that seamlessly interacts with built-in browser features.
Tune into all of the developer news
Following the keynotes, we’ll be livestreaming more sessions May 19-20. Then check out all of the Google I/O announcements and updates with over 85 sessions, codelabs, and more available on demand starting May 21.
Explore this announcement and all Google I/O 2026 updates on io.google.
Cloud
Web
AI
Cloud
Events
I/O
chrome
AI/ML
Android
Bringing Gemma 4 12B to your Laptop: Unlocking Local, Agentic Workflows with Google AI Edge
Introducing Wednesday Build Hour
Speeding Up AI: Bringing Google Colossus to PyTorch via GCSFS and Rapid Bucket
Supercharge your integration workflow with the Google Pay & Wallet Developer MCP server
Stanford’s Marin foundation model: The first fully open model developed using JAX
Get ready for Google I/O: Livestream schedule revealed