Google I/O 2026 introduced a suite of advancements focusing on frontier AI models, agentic capabilities, and integrated user experiences across its product ecosystem.

Core Models and Intelligence:
At the forefront are the latest iterations of the Gemini models. Gemini 3.5 Flash is launched as the first in a new series, combining frontier intelligence with action-oriented capabilities. It is generally available via Google Antigravity, the Gemini API, Google AI Studio, and Android Studio. Flash delivers performance comparable to larger flagship models but with significantly higher speed. It showcases superior performance on agentic and coding benchmarks, including Terminal-Bench 2.1 (76.2%), GDPval-AA (1656 Elo), and MCP Atlas (83.6%). Its efficiency makes it suitable for long-horizon agentic tasks, reducing development time and cost. Gemini 3.5 Pro is under internal use and slated for release soon. Gemini Omni is a new multimodal model capable of generating any output from any input, starting with video. It fuses Gemini's intelligence with generative media models, demonstrating advanced world understanding, multimodality, and editing. Omni integrates an intuitive comprehension of physics (e.g., gravity, kinetic energy, fluid dynamics) with knowledge of history, science, and culture, enabling photorealistic and meaningfully narrative creations. Videos generated by Omni incorporate imperceptible SynthID digital watermarks for content verification via the Gemini app, Gemini in Chrome, and Search. Omni can use any reference (image, text, video, audio) to produce a cohesive output. Gemini Omni Flash is rolling out to Google AI Plus, Pro, and Ultra subscribers through the Gemini app and Google Flow, and freely available for users (18+) in YouTube Shorts Remix and YouTube Create. It facilitates conversational video editing, allows application of cinematic effects with simple prompts, supports uploads from camera rolls, and includes built-in templates. In Google Flow, Omni Flash enhances character consistency, preserving identity and voice across scenes, and enables blending real-world inspiration with generated content.

Enhanced User Experiences with AI: AI Search has been upgraded with Gemini 3.5 Flash as the new default model, driving its "AI Mode" which has surpassed 1 billion monthly users. A redesigned, intelligent Search box allows queries using text, images, files, videos, and Chrome tabs, performing reasoning across them. The experience integrates AI Overviews and AI Mode into a seamless AI Search flow.
New information agents within Search allow users to create, customize, and manage AI agents that operate 24/7 in the background, monitoring web content (blogs, news, social posts) and real-time data (finance, shopping, sports) to provide synthesized updates and suggest actions. Generative UI in Search, powered by Google Antigravity and Gemini 3.5 Flash, dynamically builds custom interfaces based on queries, assembling interactive visuals, tables, graphs, or simulations. For ongoing tasks (e.g., wedding planning), Search can create custom experiences like dashboards or mini-apps. Personal Intelligence in AI Mode is expanding globally, allowing secure connection of apps like Gmail and Google Photos, with Google Calendar soon, prioritizing user control and transparency.
The Universal Cart introduces an intelligent shopping hub across Google platforms (Search, Gemini, YouTube, Gmail). It proactively identifies deals, price drops, tracks price history, and alerts on stock availability. Powered by Gemini models, it intelligently anticipates needs, flags product incompatibilities, suggests alternatives, and optimizes payment methods based on Google Wallet perks and loyalty information. The Universal Commerce Protocol (UCP) enables smooth in-Google checkout or direct transfer to retailer sites.

Streamlining Daily Life with the Gemini App: Gemini Spark is a 24/7 personal AI agent designed to navigate digital life and take action autonomously under user direction, built on Gemini 3.5 and Google Antigravity. It is rolling out to trusted testers and Beta to Google AI Ultra subscribers. Future features include direct interaction via text/email, custom sub-agents, and authorized payments. Daily Brief is an out-of-the-box agent that organizes and prioritizes the day by analyzing emails, calendars, and tasks, providing personalized, actionable digests.
The Gemini app features a redesigned interface, Neural Expressive, with fluid animations, vibrant colors, new typography, and haptic feedback. It transforms content presentation from text walls to interactive layouts with zoomable images, timelines, and embedded visuals. Gemini Live now opens immediately and inline, using a new, smarter model. Google Pics is a new image creation and editing tool built on the Nano Banana model, offering features like object segmentation, text editing, and Workspace integration for generating diverse visual content. Docs Live allows voice-based document creation and editing, structuring thoughts and pulling relevant details from Gmail, Drive, Chat, and the web. Similarly, Talk to Keep enables "brain dumping," converting spoken thoughts into organized notes and lists.

Agent-First Development and Creativity: Google Antigravity, an agent-first development platform, sees significant expansion with Antigravity 2.0 (standalone desktop application for orchestrating multiple agents), Antigravity CLI (command-line interface for instant agent creation), and Antigravity SDK (programmatic access to the agent harness for custom behavior and hosting). Native voice support for Gemini audio models and integration with Android, Firebase, and Google AI Studio are added. The platform emphasizes multi-agent orchestration, subagents, hooks, and asynchronous task management, significantly collapsing multi-day engineering efforts into hours. Gemini 3.5 Flash is co-optimized with the Antigravity agent harness. Google AI Studio now facilitates building native Android apps directly, with integration to Google Play Console for direct publishing to test tracks. Developers can deploy their first two apps to Google Cloud for free. It also supports exporting projects to Google Antigravity for local development. Managed Agents in the Gemini API provision remote Linux environments where Antigravity agents can reason, plan, call tools, execute code in isolated sandboxes, and browse the web, powered by the new Antigravity agent built with Gemini 3.5 Flash.
New open web standards like WebMCP allow exposing structured tools to browser-based agents, while Modern Web Guidance provides expert-vetted skills for AI coding tools. Chrome DevTools for agents gives AI agents visibility for code verification, debugging, and optimization.
A new $100 AI Ultra plan is introduced for developers and advanced users, offering higher usage limits, cloud storage, and enhanced features. Google AI Pro subscriptions now include YouTube Premium Lite.

Productivity and Creativity Enhancements: AI Inbox in Gmail, previously for Ultra subscribers, is rolling out to Plus and Pro users. It now generates personalized draft replies, surfaces relevant Google Docs/Sheets/Slides links, and offers improved task management. Gmail Live will allow conversational queries within the inbox. Google Flow, an AI creative studio, now incorporates Gemini Omni Flash and introduces Google Flow Agent, a multi-step agent capable of planning and reasoning for complex creative tasks (e.g., brainstorming, batch editing, asset organization). Google Flow Tools enable users to create custom creative tools and workflows using natural language, which can be shared and remixed. Google Flow Music allows conversational direction of shareable music videos with granular refinement capabilities via Gemini Omni. Pomelli and Stitch also see updates for brand content and real-time design collaboration.

Scientific Breakthroughs and Learning: Gemini for Science is a collection of experimental tools on Google Labs. Hypothesis Generation (with Co-Scientist) simulates the scientific method using multi-agent "idea tournaments" for hypothesis evaluation. Computational Discovery (with AlphaEvolve and Empirical Research Assistance) is an agentic research engine that generates and scores code variations for complex modeling. Literature Insights (with NotebookLM) searches scientific literature, structures results into tables, and allows chat-based analysis. Science Skills integrates insights from over 30 major life science databases (e.g., UniProt, AlphaFold Database) into agent-first platforms like Google Antigravity for complex scientific workflows. Experimental tools like Paper Assistant Tool (PAT) and ScholarPeer are being developed for agentic peer review and scientific validation. Ask YouTube is a new conversational search experience that compiles relevant videos (long-form and Shorts) and provides interactive, structured responses to complex queries. Android XR is advancing with intelligent eyewear: audio glasses (partnerships with Gentle Monster, Warby Parker, Samsung) and display glasses, compatible with Android and iOS devices.

Safety and Content Authenticity: SynthID, Google's imperceptible watermarking technology, has expanded its verification capabilities for image, video, and audio to the Gemini app (already 50 million uses globally), Search, and Chrome. Users can inquire about content origin (e.g., "Is this made with AI?"). Integration with C2PA Content Credentials is also rolling out, allowing verification of content alteration and tools used.