Gemini 2.0 Flash Experimental lands on Android

Google has been rolling out some exciting experimental updates for Gemini this week, with Gemini 2.0 Flash landing on Android to go alongside the desktop version that’s been on the web for the past week or so. If you haven’t yet, you should go read about how big of an update Gemini 2.0 is on the whole. It’s a big deal.

Last week, Gemini Advanced subscribers got access to Gemini 2.0 Flash Experimental on the web, accessible via the version picker on desktop and mobile web. This new model promises significant performance improvements for complex tasks like coding, math, reasoning, and instruction following. Whether you are tackling complex coding challenges, crunching numbers for projects, or creating detailed instructions for business plans, this model is designed to handle complex tasks with greater ease.

Keep in mind that this is an early preview, so Google warns that it might not always work perfectly and lacks real-time information access and compatibility with some Gemini features. We’re seeing the new version in our accounts here at Chrome Unboxed, so it seems to be widely rolled out at this point and should be available for most of you to begin testing.

Gemini 2.0 Flash on Android

Today’s news is about Gemini 2.0 Flash now being available on Android devices. Open your Gemini app on Android, and you should see the “2.0 Flash Experimental” option in the model dropdown at the top. Bear in mind, you’ll need to be a Gemini Advanced user and have the latest Google app installed to use it. With Gemini 2.0 Flash is all about speed and performance; here’s how Google described it at launch:

Introducing Gemini 2.0 Flash as an experimental model in Gemini

What: Gemini users now have access to our latest experimental model, a chat optimized version of 2.0 Flash Experimental with improvements on a number of key academic benchmarks and speed. This experimental model is meant to be an early preview and can have unexpected behaviors and may make mistakes. Additionally, some Gemini features won’t be compatible with this model in its experimental state. You can access this experimental model in the Gemini model drop-down.

Why: We believe in rapid iteration and bringing the best of Gemini to the world. Which is why we want to give users access to cutting-edge models with our latest enhancements. Your feedback helps us improve these models over time and learning from experimental launches informs how we release models more widely.

So far, I feel like talking with Gemini 2.0 Flash is a bit quicker, but that could 100% be placebo. There’s no doubt that this model will outperform previous iterations of Gemini in larger tasks, but until you really start to push it, there’s no real benchmark for this sort of thing. But between the new Veo 2, Imagen 3, Whisk and this latest Gemini update, I’m a little blown away with the AI products we’re seeing from Google right now. And maybe a tad bit scared.

This is all moving very fast right now, and the innovation levels are pretty staggering. One of the things I like most about all these AI advancements is the fact that Google continues to keep continuity between the web and Android, not making one feel more important than the other. And with AI generally living in the cloud for the bigger stuff, that just makes sense. There’s a lot to try out right now in Google’s AI arsenal, so don’t hang here too long. Go give it all a test and see what you think!