Gemini 3.5 Flash reaches GA as Google turns Search into an agent surface
Original: Watch 12 Google I/O 2026 keynote videos of the top announcements and updates View original →
Google’s I/O 2026 AI push is less about one chatbot and more about putting Gemini into every surface where work starts. In a May 28, 2026 recap, Google highlighted the general availability of Gemini 3.5 Flash, the rollout of Gemini Omni Flash, and a new Search agent layer.
Gemini 3.5 Flash is the first release in the Gemini 3.5 family. Google describes it as a model for agents and coding, built for complex long-horizon tasks with real-world utility. The distribution is unusually broad: Google Antigravity, the Gemini API in Google AI Studio, Android Studio, Gemini Enterprise Agent Platform, Gemini Enterprise, AI Mode in Search, and the Gemini app. Gemini 3.5 Pro is being used internally and is slated for rollout next month.
Gemini Omni is the multimodal part of the same strategy. Google says the model family can combine images, audio, video, and text as input, generate high-quality videos grounded in Gemini’s real-world knowledge, and support conversational video editing. The first model, Gemini Omni Flash, is rolling out globally to Google AI Plus, Pro, and Ultra subscribers through the Gemini app and Google Flow. It is also rolling out at no cost to users on YouTube Shorts and the YouTube Create App.
The Search changes may matter most for distribution. Google described information agents that users can create inside Search by adding “keep me updated” to a query. Those agents will reason across the web, blogs, news sites, social posts, and fresh data such as finance, shopping, and sports, then send updates when relevant. The feature starts this summer for Google AI Pro and Ultra subscribers.
The competitive point is clear: Google is not only shipping models, it is attaching agent behavior to default user surfaces. If the rollout works, Gemini’s advantage will be measured not just by benchmarks, but by how often users can move from question to action inside Search, an IDE, a video tool, or an enterprise workflow without switching products.
Related Articles
Google has put Gemini Embedding 2 into public preview through the Gemini API and Vertex AI. The model is Google’s first natively multimodal embedding system, combining text, image, video, audio, and document inputs in one embedding space.
Google AI Studio promoted Gemini Embedding 2 in a March 12, 2026 X post, and Google’s March 10 blog post says the model maps text, images, video, audio, and documents into a single embedding space. Google says it is in public preview through the Gemini API and Vertex AI and is designed for multimodal retrieval and classification.
Google on Mar 17, 2026 introduced new Gemini API features for agentic workflows, including combined built-in and custom tools, context circulation across tool calls, and Maps grounding for Gemini 3. The update is designed to reduce orchestration work for complex multi-step applications.