On 12/12, in addition to OpenAI’s announcement that ChatGPT is being integrated into Apple’s systems, Google also announced the launch of its latest AI model, Gemini 2.0. Google stated that Gemini 2.0 leverages new multimodal capabilities to build new AI agents, realizing the vision of a universal assistant. Reading Google’s description of Gemini 2.0 alone might feel a bit abstract, so today we’ve compiled several practical applications that use Gemini 2.0 to show everyone what Google means by AI agents.
Real-world applications of Google’s latest AI model Gemini 2.0, acting as a piano teacher and game assistant.
Google announced today the launch of its next-generation AI model, Gemini 2.0. Google stated that Gemini 2.0 can build new AI agents through multimodal capabilities such as native image and audio output, as well as built-in tool usage. Gemini 2.0 is expected to include multiple models, and the newly released Gemini 2.0 Flash Experimental features low latency and optimized performance. As for the performance test results of Gemini 2.0 Flash Experimental, everyone canGo read the related reports.This is mainly to introduce what you can do with the Gemini 2.0 Flash model.
According to the Gemini 2.0 Flash application shared by Google, Gemini 2.0 Flash can become your gaming assistant. While you’re playing, Gemini 2.0 Flash will read your screen in real time and offer suggestions. For example, when you’re playing Clash of Clans, you can ask Gemini 2.0 where you should attack from, and Gemini 2.0 will give you attack position recommendations based on what it actually sees on screen. So does this mean that from now on you don’t need to open up guides when playing games — Gemini 2.0 can just be my guide assistant?
In addition to the official examples, netizens @tsyn18 Using Gemini 2.0 Flash’s AI feature “Stream Realtime,” Gemini 2.0 can play games with netizens while recognizing the screen, leading netizens to marvel that the era of working and playing games alongside AI is about to arrive.
きたきたきた！！！！！
これだよこれ！これがずっっっとやりたかったんだ！！Googleの最新AI機能「Gemini 2.0 Flash」の「Stream Realtime」を使うことで、AIに『画面認識』させながら一緒にゲームのプレイができました！… https://t.co/7grTyunrv4 pic.twitter.com/QRNEIrSZOY
— 中村俊也｜ＡＩキャンプ® (@tsyn18) December 12, 2024
As mentioned earlier, Gemini 2.0 can handle multimodal inputs such as images, videos, and audio, and supports multimodal outputs, generating text and native images simultaneously, as well as multilingual text-to-speech (TTS) functionality. According to Google’s official applications, Gemini 2.0 can help you read your emails and find and record the content you ask for. For example, when you receive a booking notification, Gemini 2.0 can help you find and remember the most important room password from a long check-in notification.
Many people can’t understand what the labels on their clothes mean when doing laundry, and having to look them up online every time is a hassle. But with Gemini 2.0, you can simply use your phone camera to scan the care label on your clothes, and Gemini 2.0 will tell you how that garment should be washed. 
According to netizen tests, Gemini 2.0 can analyze in real time what your phone camera sees, and by identifying boat types, terrain, and popular statistics, find which small island in Thailand the user is on.
It’s incredible how Gemini 2.0, in real-time, analyzed what I was seeing on my phone and figured out which small island in Thailand I’m on, just by recognizing the type of boats, the terrain, and popularity statistics. We’re just months or a few years away from HER—I feel a bit… pic.twitter.com/diTRq7kr9z
— Victoriano Izquierdo (@victorianoi) December 12, 2024
In addition to identifying your location from the background and objects it sees, Gemini 2.0, when paired with a camera, can also become your real-time piano teacher. For example, it can tell you which keys to press and where, while seeing your perspective through the camera and engaging in real-time conversation with you.
Gemini 2.0 acts as my virtual real-time piano teacher.
This new Google AI model can see your screen and have live conversations with you.
Unfortunately, it can’t “hear” songs yet. pic.twitter.com/DGdwKriIKC
— Kevin Kern (@kregenrek) December 12, 2024
What are your thoughts after reading the above sharing on Google Gemini 2.0 applications? Regular users can now also find the chat-optimized version of Gemini 2.0 Flash Experimental in the model dropdown menu on Gemini’s desktop and mobile web versions. However, features like text-to-speech, native image generation, and the app will still require a bit more waiting.
For those interested in the optimization of the Gemini 2.0 Flash Experimental model and the differences in upcoming future models, you can visit the link below to read the related report:
Source: KOCPC Chinese