You Can Try These New Google I/O Features for Free Right Now

Google announced a plot of updates at I/O 2026. The company’s first modern smart glasses (dubbed “Intelligent Glasses) will arrive later this year, as will Google’s Universal Cart shopping assistant. There’s even a brand new AI model, Omni, that can create any output from any input, which should do wonders for our AI-generated misinformation problem. But not everything Google announced Tuesday is available now (or for free, for that matter). Most new Workspace features, for example, require some type of subscription, as do many upcoming AI improvements. If you want to test Omni today, you can, but you have to pay for one of Google’s AI subscriptions.

The good news, however, is that there are some features and changes Google announced during I/O that you can try right now, without paying a dime. These are new tweaks to apps and services you may use every day, like Google Search, the Gemini app, or Gemini itself. Although Google is rolling out more free I/O features later this year (and starting this summer), here are the announcements you can take advantage of today:

Gemini 3.5 Flash is now in Google Search and the Gemini app

Google I/O 2026 was all about AI. As such, one of the biggest announcements from the keynote was Gemini 3.5 Flash, the latest update to Google’s Gemini model. This is the first model in the Gemini 3.5 family, as Google says it is still working on Gemini 3.5 Pro.

According to Google, Gemini 3.5 Flash competes with other flagship AI models “across multiple dimensions.” The company claims that 3.5 Flash outperforms 3.1 Pro in coding and agent testing, and is an industry leader in multimodal understanding. Since it is a “Flash” model, it is designed to be faster than other models that prioritize performance over speed. At this point, Google claims that Flash 3.5 is four times faster than other “frontier” models.

Taken at face value, Gemini 3.5 Flash is an option for almost anyone using AI. If you’re a developer, Google says Flash 3.5 will save you time and money because it’s often less than half the cost of comparable models. But for the rest of us who don’t use AI to code or develop, Gemini 3.5 will be more accessible in Google Search and the Gemini app. As of Tuesday, this is the model that powers Google’s flagship AI products. Therefore, if you have used Gemini in the last 24 hours, you have probably used Flash 3.5.

Google’s new ‘Smart Search’ AI box is already rolling out

There are plenty of changes coming to Google Search, many of which fundamentally alter how the platform works. Google wants search to be a true AI-driven experience, focusing more on interacting with AI mode and AI insights than on the individual links and sources where those answers come from. As sites continue to lose traffic to these AI updates, it’s no exaggeration to say that this new approach could change the web forever.

These major changes will arrive next week. Today, however, you’ll notice other changes when you use Google Search. Of course, you now know that the underlying AI model that powers AI Mode and AI Previews is Gemini 3.5 Flash, but that’s far from the only change. Perhaps the most notable update is the new “Smart Search” box. When you select “AI Mode” in the search box, Google leverages Gemini 3.5 Flash to provide AI-based suggestions as you type. As advertised, the feature seems intended for more conversational searches rather than quick queries. Things like “I’m looking for a new hobby and interested in pottery.” As you type, Google may suggest the following: “Is wheel throwing or hand building easier to learn?” » The idea is to guess what you’re going to ask to save time, or to suggest search terms that you might not have thought of yourself. The feature is also multimodal, so you can search with images, files, videos, or Chrome tabs, in addition to text.

Google started rolling out the feature to users on Tuesday, so it might be available on your end, but it may take a while. (At the time of this article, I haven’t seen it yet.)

AI mode just got easier to use

AI previews are divisive: while many may find them handy for getting quick results, others are critical of the tool’s accuracy. (Remember, AI Overviews once recommended putting glue on pizza to keep the cheese from falling.) However, if you like using Google AI search tools, you might like this next change.

On Tuesday, Google rolled out an update to AI Previews that makes it easier to switch directly to AI mode from the results. Now when you get an AI preview, you can choose to expand it, which has a chat box at the bottom of the window. You can use this to engage in a “conversation” via AI mode, if you want to continue asking questions on the topic.

More search features coming this summer

Google is also offering many new search features later this summer. Google AI Pro and Ultra subscribers will be able to try “Information Agents,” which keep tabs on certain topics, sales, or trends and alert you when things change. Free and paid users will be able to use agents to book things, including restaurant and private karaoke room reservations, right in search. Additionally, Google is rolling out Universal Cart to everyone this summer, which lets you add items from multiple stores into a single digital cart.

But perhaps the biggest update for free users, in my opinion, is agent coding. Google will use Gemini 3.5 Flash to create interactive elements in real time based on your queries. If you’re asking about black holes, Google can create a demo for you to play with to see how they work. These will also be rolling out for free this summer.

The Gemini app received a “Neural Expressive” design update

In addition to running Gemini 3.5 Flash, the Gemini app is getting a new visual design refresh. Google calls it “Neural Expressive,” with new animations, colors, typography and haptic feedback. It’s definitely different in appearance than the original Gemini app, or Google’s usual design scheme, so if you’re interested in UI updates, you can check it out today.