It's been an extraordinary year since the last I/O, and Google came into this one with a clear mandate: move past the model arms race and show real value in products people use every day. The 2026 keynote delivered on that in a big way, with Gemini touching almost every surface Google owns.
Sundar Pichai opened with figures that frame the scope of what Google is building toward. The Gemini app now has 900 million monthly active users, roughly doubling from 400 million a year ago. Two years ago, Google was processing 9.7 trillion tokens a month. Last year that grew to 480 trillion. Today, that number has jumped seven times to 3.2 quadrillion tokens per month.
Alphabet has committed between $175 billion and $185 billion in capital expenditures for 2026, nearly double the $91.4 billion it spent in 2025, with the bulk directed toward AI compute infrastructure. That's the financial backdrop for everything announced on stage.
Gemini 3.5 Flash and Omni: Two New Models
Google introduced Gemini 3.5 Flash, the first in a series of models combining frontier intelligence with action. Compared to 3.1 Pro, 3.5 Flash is better across almost all benchmarks, with huge progress in coding and an extraordinary jump in GDPVal.
It surpasses 3.1 Pro in coding, agentic, and multimodal benchmarks, with the cost and speed of the Flash series at 4x faster than other frontier models in terms of output tokens per second. It's rolling out today in the Gemini app, Search, Antigravity 2.0, and the Gemini API. Gemini 3.5 Pro is currently in testing and will be available next month.
Alongside 3.5, Google released Gemini Omni. Omni can create anything from any input, starting with video, and represents a leap forward in world understanding, multimodality, and editing. Gemini Omni Flash accepts image, audio, video, and text input and outputs video grounded in real-world knowledge that can be easily edited.
Gemini Spark: A 24/7 Personal AI Agent
This is the announcement that changes the most about how people will interact with Google's AI going forward. Gemini Spark is described as "your personal agent" that takes actions on your behalf. It integrates with Gmail, Docs, and other Google Workspace apps before expanding to other third-party tools via MCP over the summer.
Gemini Spark will be available next week to Google AI Ultra subscribers in the US. Pichai specifically noted that when using Gemini Spark, "yes, you can close your laptop." That's a direct shot at the agentic workflows that have made Claude and other tools popular with developers.
