Asking AI and expecting it to respond sanely is no longer enough: we need do it quickly, make it efficient and can handle large amounts of information. This is exactly what Google promises with its Gemini 2.0, the new model is already in testing. And I had access from Android, it's a notable improvement over Gemini 1.5 Pro and it almost completely banishes Google Assistant.
That generative AI is here to stay is indisputable, just look at the shift towards artificial intelligence that all the big tech companies have taken. Including smartphone makers, who each have (or will have) their own suite of AI tools; many of them are based on Google's solution, Gemini.
Gemini 2.0 is now included in the Google One AI subscription
Gemini language models selectable in the browser
With the popularity of Open AI and its ChatGPT, Google has had to devote a lot of its resources to focusing its efforts on AI. towards generative economic models. The company has developed many tools with which these models are trained, with one drawback: until the arrival of Open AI, Google did not discover the enormous potential of LLMs (long Language Models) for users individual. Now press the accelerator to recover positions.
Google recently released its Gemini 2.0. And it's now available in the mobile app and on the web, although only for those of us who have Gemini Advanced. If this is your case, you will see that there is a small selector in the upper left corner: click there and select the new model, Gemini 2.0. Remember: it's still in testing, so it may be more inaccurate than usual.
Gemini 2.0 is in the experimental phase. Responses should be taken with great caution.
And what is the main difference I noticed compared to Gemini 1.5 Pro and Gemini 1.5? The response speed is higher, argue better what he explainsI found it more reliable and, according to my impressions, there is an effort towards naturalness. A little less robotic, but with the same desire for expansion: despite the fact that I told it to be concise, Gemini 2.0 is appreciated as a speaker.
Gemini 2.0 Flash in One Google Pixel 9 Pro
Google has included the new version of its multimodal model, but with differences:
- Gemini 2.0 Flash, the cleanest model, is available in the Android app.
- On the website, you can select Gemini 2.0 Flash and Gemini 2.0 Advanced, the model with the greatest capabilities. Both are still in beta.
- If you have an iPhone, you don't have to try Gemini 2.0 in the app: it's only found on Android. Yes, you can access the experiment from the web, in both versions: Flash and Advanced.
Both are in the experimental phase and are not yet fully capable: They do not support document uploadingYes, photos. Additionally, the new Advanced model doesn't use real-time searches, so it only responds based on your training. Flash creates photos with image 3, Advanced does not have this option yet.
Noticeable improvement in speed, reasoning and argumentation. It's still not concise
Chatting with Gemini 2.0 Flash on a Google Pixel 9 Pro
As is often the case with every new software release, Gemini 2.0 promises to improve performance and capabilities. In my experience there is a change, although not exaggerated. Yeah, It's faster and I like that it's more precisebut it's not a giant step either.
Google took very little time to include the new language models in its particular applications and without leaving us Europeans aside, a positive point. Working with Gemini 2.0 provides greater reliability. And now all that remains is for Google to take it out of the beta version to expand its use and the associated tools, in particular all those which allow you to interact with the phone. Google Assistant has nothing left to do.
Cover image | Edited Gemini Image 3
In Xataka Mobile | Movistar now allows you to configure your eSIM for free in store if you meet a condition
In Xataka Mobile | What to do if you see advertisements and advertisements on your mobile phone
