Android

How to improve the sound of your Android mobile by just activating a hidden option

by

Let’s tell you How to improve the sound of your Android mobile activating the HD audio option. This is an option that is somewhat hidden, and that will work for both the audio of the phone and for any compatible device that we connect, such as headphones. Here, remember that although the option is available by default on Android, The mobile will have to be compatible With the sound in high definition to be able to activate it, and use it whenever it is available. If it is not compatible, the option will appear but cannot be activated. Improves Android’s sound with HD Audio The first thing you have to do is Activate developer options. For that, in the Android settings enter the section Device information either of the phone. Once inside, click several times where it puts Compilation number. They are usually seven times, but click until I tell you that developer options have been activated. Now, you have to click on Developer options. You can find it directly in the adjustments, or within the section of System. It will depend on your mobile and manufacturer. Now, within the developer options Search and activate the option HD audio. If you do not find it right away, use the search button you have up to the right. The option can only be activated if your mobile is compatiblethat is, you will not always be able to do it, it will depend on your device. In Xataka Basics | Android 16: 17 Functions and some trick of the new version of Google’s mobile operating system

What are they, how these notices are configured and how android are configured

by

Let’s explain What are they and how to configure Google earthquake alertsa security system that you can configure on Android. In this way, if there were an earthquake near where you live you will have an extra notice of Google itself beyond that you can send the institutions of your country. We are going to start this article explaining what this type of Android mobile notices work. Then, we will briefly explain how you can access your configuration to verify that they are notified. What are alerts for earthquakes and how This type of alerts is part of the Google seismic detection system, Shakealert. It is a system that uses a series of seismic sensors in some parts, but also Telephones of the users themselves to detect earthquake movements. One of the sensors that is present in all mobiles is the accelerometer, which is used to detect its movement in several applications and functions. Android also uses this sensor to detect movement changes that can be interpreted as an earthquake for its internal algorithms. When your mobile detects this type of movements, He sends a signal to Google servers. When it does, Google has internal systems that combine your information with which other phones in the area send, and thus try to find out if an earthquake is taking place in that area. In the event that Google detects that there are indeed many mobiles by sending this type of warning signals in specific areas, and that it seems to be an earthquake, then Send an alert to Android users through a notification. The alerts that are sent can be in two ways, depending on the intensity of the detected earthquake. If it is 4.5 degrees in MMI scale (No Richter) or higher, then a strong sound alarm is sent that ignores the ways of not disturbing, and turns on the screen of your mobile. In the case of being somewhat lower, then a normal notification is sent. How to configure these alerts To check if you have this function activated, you have to go to the settings of your Android mobile. Inside, click on the section Personal Security and Emergencies That will appear almost down at all. Once here inside, click on the option of Seíncos alert. With this, you can verify that these alerts are activated, and that you will receive them in the event that an event is detected near you. Here, remember that the configuration can be somewhere else or have another name depending on the manufacturer of your mobile. Therefore, you can use the settings search engine and look for words We are either earthquakeand the adjustments used by these terms will be located without having to look for them by hand. In Xataka Basics | National Alert Network Viewer (RAN): What is and how it works to see weather or earthquakes notices

Android does not have to be an absolute disaster in updates. Samsung has had to demonstrate it

by

Just a month ago Google updated its last pixel to the latest stable version of Android, 16. Today, in its presentation event of the Samsung Galaxy Z Fold7, Galaxy Z Flip7 and Galaxy Z Flip7 Fe, the company has presumed something unusual: the three have the latest version of Android. It is the two very clear points test in 2025. Android 16. That a manufacturer throws a phone with the latest version of Android just a month of its launch is very strange. Manufacturers usually take months and months to refine the ROM to the latest version, especially in cases such as Samsung, with a heavy customization layer. Laning three phones to the market with Android 16, when the Pixel are starting their journey with this version is a clear message: Samsung wants to be the first, and not only in speed. The seven years of updates. Samsung was the first to achieve something unusual: to advance Google in update policy. The manufacturer offers seven years of system updates and another seven years of security patches. Only Google (who arrived after Samsung) and Honor, who has joined the car, manage to offer their customers these figures on their best phones. Seven years of support are major words. It is not just the proof that Android phones They can update during the same time as an iPhoneIt is a trusted proof for the brand with respect to its own product. Leader in AI. Samsung has a Google partner to enhance his suite of artificial intelligence functions and, curiously, his Galaxy AI It is the most complete the AI ​​ecosystem in the mobile market. Above Google’s own proposal in his Pixel and far ahead of what rivals like Apple are doing. Samsung did not want to implement Gemini Nano To make the minimum effort: it has integrated its functions with the native apps of One UI, to the point where they are practically indistinguishable from any usual adjustment. Just press the AI ​​button to take a step further apps such as: Gallery Phone Circle to Search (compatible with S-PEN) Voice recorder Translator GENERATION OF WINDERS Third -party apps, such as WhatsApp and keyboard translation The lessons. For years, Touchwiz was the ugly duckling of the Roms on Android. Worked slow (there was lag) I was very overloaded with functions At the aesthetic level it was … debatable Today, it has one of the most refined ROMs at the interface level, it is able to manage all its native functions and apps without slowing down (even in mid -range and entrance mobiles) and is marking a path as a reference in updates and Ia. Image | Samsung In Xataka | The mobile the AI ​​promises, but I only see repeated tricks. The real ace under the sleeve is called “agent” and comes on the way

How to use Gemini to summarize YouTube videos or ask questions about its Android content

by

Let’s explain How to summarize YouTube videos or ask questions about them using Gemini on Android. Both artificial intelligence and the video platform belong to Google, so they get along very well between them, and it is easy to interact. With this simple method, when you are watching a video on your mobile and do not have time to see it whole, you can ask Gemini a summary or ask specific questions. For example, if it is a recipe you can ask you to remind you of the amount of salt you should throw, and if it is a video of tricks you can ask you for a concrete one or ask any other type of question. To be able to use this trick, Gemini must be your main assistant on your mobile. If it has not been assigned as a default assistant, you can follow the steps of our article on How to change the Google Assistant for Gemini On your Android mobile. Ask Gemini about a YouTube video The first thing you have to do is open YouTube and be reproducing the video on which you want to ask the question. So, invokes Google Gemini. The AI ​​will recognize that you are watching a YouTube video, and above the writing field will show you two shortcuts. Here, you have to click on the option Ask questions about this video that will appear in the emerging window. When you choose that option, Gemini will add the URL of the YouTube video, and will allow you to write or talk to ask the question you want. Here, when the window is already with the added video, Ask him to make you a summary about the videoeither writing the request with natural language or using the option to speak. Then, directly in the emerging window you will see A textual summary of the videoincluding points of interest and concrete moments in which certain things are said. The summary will be short, but explaining the content well. You can also ask other specific questions related to the content of the video. Gemini will be able to analyze everything that is said in it, and will answer you to any questions you ask. Therefore, You no longer need to see the whole video To get the answers, you can save time in long videos with this function. In Xataka Basics | The best PROMPTS to save working hours and do your homework with Chatgpt, Gemini, Copilot or other artificial intelligence

The United States is creating “the iPhone of AI”. China is doing “the Android of AI”

by

There is an interesting paradox in the current development of AI that says more about geopolitics than about technology: USAtraditionally leader in proprietary software and monetization, is building closed models of AI. Chinahistorically more restrictive with the flow of information, it is leading the open source AI. An investment that is not accidental: each block adopts the strategy that best serves its structural interests. → United States (OpenAI, Anthropic, Google), is building the equivalent of the IPhone of the AI: Very sophisticated systems. Vertical integration. PUPLE POLIDED PREMIUM EXPERIENCES that justify charging for access. The logic is clear: when you control the best GPUS, the main cloud services and you have the capital to train the most advanced models, monetizing that advantage through payment APIs makes immediate economic sense. It is the model that has holding American software for decades. → China, on the other hand, with Deepseek, Qwen and Minimax Mastering the rankings of open models, it is creating the Android of the AI: A free and modifiable ecosystem. Downloadable for local use without depending on payment APIS. Permissive licenses without business restrictions, Unlike Meta with Flame. Accessible source code for independent research and development. As we said at the timethis approach is not due to altruism, nor by cultural difference. It is a strategy based on creating global dependence on a Stack Chinese technological when American becomes inaccessible. Or at least, to present it as a very attractive alternative even before seeing its results. It is true that he calls is American and is also in the league of the greats being open source (With an asterisk of the size of Mestalla). A habitual hypothesis also seems feasible: that target was released precisely because it was behind. The classic defensive strategy to erode the advantage of leaders. China does it from another position: growing strength. And in addition, having to deal with access restrictions to western markets. When your rival controls chips and cloud platforms, and can cut the tap at any time, the only way to create an alternative ecosystem is to make it so accessible that the world cannot ignore it. Soft Power Technological: give today to dominate tomorrow. That, or cling to developing markets plus the gigantic domestic market. This is how Huawei stopped competing to build a parallel reality. The numbers speak alone. According to the Benchmarks of artificial analysis (based in California, not in Shenzen), The three best open source models leave China. Each startup that Depseek uses instead of GPT, each country of the southern hemisphere that displays Chinese models because they are free, each university that trains over Qwen instead of Claude, is another node of the ecosystem that the United States cannot censor, regulate or disconnect unilaterally. It is a story very similar to that of The slow but unstoppable Chinese independence of GPS: In two decades it is no longer that I do not need it, it is that Beidou has reached 140 countries. It is also true that There are traps in both strategies: The American model generates immediate income but creates incentives for the rest of the world to look for alternatives, especially when commercial wars are intensified. Competence. The Chinese model conquers users but at some point it will need to monetize without frightening its base. The latter is something that Google learned with Android for the bad ones: after conquering 70% of the world, it began to monetize it aggressively and ended up receiving huge fines in Europe, antitrust demands in the United States and Chinese alternatives … as Harmonyos de Huawei. The circle closes. The lesson is that giving technology to create dependence works, but monetize that dependence once established attracts the attention that nobody wants: that of regulators. It is the dilemma that China will eventually face its “free” models of AI. The real battle is not for today’s best, but to control the Intellectual infrastructure Tomorrow. The United States sells the premium service. China gives the universal operating system. And the outcome will be that of two historically successful, but opposite philosophies: the American persistence by aggressively monetizing in front of Chinese patience to invest decades if necessary until it can lead. Immediate income against long -term dependence. Time will say. In Xataka | Four AI companies are monopolizing the intellectual future of humanity. They are not good news Outstanding image | Ilgmyzin in UnspashAlibaba, Xataka

China is flooding from devices without a trace of Android. Huawei is the main responsible

by

Huawei’s ability to respond to the numerous restrictions imposed by the United States government is being the best example. The company is in the process of continuing to it to go beyond the Kirin 9020marking rhythm of launches with The Pure 80 Ultra and its Advanced Chamber Systemand achieving what no other manufacturer had achieved to date: create a real alternative to Android. Huawei is working for the plan. Harmonyos It is a strategic turn towards the technological self -sufficiency that China has been chasing for decades. An own operating system, with national code and without a trace of Android, the platform that Huawei used in the initial phases of the system. At the end of 2024, Huawei’s Annual Results Report He presumed that more than one billion devices had Harmonyos. In this first half of 2025, we have some more specific data. At all rag. According to Canals, Huawei has sent a total 103 million smartphones with Harmonyos and 21 million tablets from 2021 at the end of 2024. Of those 103 million, almost half (46 million) were sent last year, showing how the last two years are being key to the growth of the company. The data, without context, are only data. To understand at what point Huawei is located to know that, in 2021, the estimates of sale of smartphones gave Huawei 10 million. The figure is important and shows an upward trend: Huawei has gone from the free fall that suffered between 2021 and 2022 to a sustained recovery process in the last two years. In 2022 the figure doubled to almost 20, to end up arriving at 2025 at 46 million. More prohibitions, more shipments. The United States Plan to isolate Huawei in the world has a double edge. The company has collapsed in Europe and left its position to its sister honorbut the efforts to grow nationwide are clearer than ever. Huawei is triumphing in China, and the consultants point out that it will be one of the mobile companies that will grow the most global in 2025 (growth marked by the path they are following in China). A plan that will accelerate with the license restrictions that affect its PCs. Huawei can no longer sell Windows computers: the first solution is Patch existing ones with Linuxthe second, Sell ​​your own computers with Harmonyos. The dream, getting closer. Huawei has made restrictions the greatest opportunity for growth in its history. An approach towards A device with 100% national components, A universal software ecosystem (mobile, tablet, pc, car, wearables) and, most importantly, The only possible and alternative rival to Android out of Apple and his proposal with iOS. Huawei is drawing an alternative scenario in which every Chinese manufacturer who wants to disconnect from the United States can have Harmonyos alternative. It is something that begins to be seen in industries such as that of automotive: manufacturers like Toyota are so far back in software that The only possible solution is to bet on alternatives such as Huawei or Xiaomi. For the moment, Xiaomi, Vivo, Oppo and even honor continue to see in Europe one of its main sales pillars. But China prepares to lead the technological worldand Harmonyos will play a fundamental role in their conquest. Image | Xataka In Xataka | Harmonyos shoots in China and already surpasses iOS. The following objective is global

imitate what android had already done

by

When Apple presented Apple Intelligence I was quite confusing. One of the companies referring to the world of technology was shouting at the four winds that had practically nothing relevant on the table. One of the most striking novelties, that of A Siri with contextual understanding And capable of executing local Apple models in a conversational context, it ended up delaying without a date in view. The rest of the functions? Warm and little innovative. Meanwhile, Samsung pushed from Google’s hand to create A AI tool suite for mobile to which nothing could not be reproached. Something they recently achieved with the family Samsung Galaxy S25the maximum current reference in artificial intelligence. Apple has shown how AI for mobile. Even missing a lot of what they promised in their daywe already know how Apple understands the AI ​​for mobiles. Its approximation is local, Leaving third parties like OpenAi The complex process (consultations that require search in Siri, contextual analysis of the screen), and centralizing in its own model the native functions of AI that execute their apps (translations, call filters, minor editions of image). It is an AI in which developers will have a lot to do: their models will be accessible so that applications can use artificial intelligence locally and without connection. But this local approach is not enough, Apple has failed to contribute a single function that was no longer present on Android. Arrive late when the rest arrives soon and well. Apple is usually said that it is late, but it is well. The problem is to be late and well when your rivals arrive soon and well. The functions presented yesterday by Apple have been present at Android for some time. The translation of real -time calls was already present in One UI. Text translation on screen is a native Gemini function that already implement manufacturers such as OnePlus and Samsung. The contextual analysis of the screen is a function that Google premiered with “Roda to search”and that has currently advanced to the point where Gemini can talk with us in real time about what he sees on our mobile. Intelligent summaries also have been hand in hand with Gemini. Apple’s approach, in functions such as spam detection, is enough less fine than Google’s. The company analyzes by means of patterns in the call to determine whether or not spam and warns us on the screen of the potential risk. Apple has opted to block all calls from unknown numbers and force the interlocutor to give explanations about who he is and what he wants. Good to block fraudulent calls, annoying for important emergencies and calls from numbers that we do not have stored on the agenda. Samsung and local execution. One of Apple’s most applauded points and its AI is the local approach. Here they seem to be in the lead, although more for demerit of the rest of their rivals than by the operation of their models. Currently, the AI ​​on Android is divided into two: Samsung and the rest of the manufacturers. The rest of Oems bet on implementing the functions of Gemini Nano With little care, so the Internet connection is mandatory. However, manufacturers such as Samsung have been allowing users for two years to use local. Advanced functions such as image generation do not work, but the elimination of sound, light generative editions (allowance) or transcripts work at home. For Apple is not what, is how. Apple has always been pragmatic. He knows that his models are more than valid for the execution of simple tasks, and delegates the complex tasks in OpenAi. It also does it, relying on three pillars. It is optional. It is private. It is interchangeable. For Apple, Private Cloud Compute It is fundamental. The company has designed an architecture that, in case of needing Internet connection for complex tasks, is extremely careful with the data launched to the server. This data shipping is encrypted from end to end, and not even company employees have access, according to Apple. A local, private and jealous with the data is a more limited AI, and Apple is willing to deal with it. Image | Xataka In Xataka | Apple has decided not to lead the AI ​​revolution. Instead, he has chosen to be his best host

Going down the price of Google Pixel 9 Pro XL, one of the best mobiles with Android

by

If you are thinking about renewing your old mobile and want to move on to a high -end and with pure Android, the Google Pixel 9 Pro XL It is the terminal that I would recommend right now. In addition, you can now take it cheaper. In Amazon, now, it has a 25% discount and is available for 897 euros. This price is for the 128 GB version, although if you want more capacity and, in addition, you are a partner of Mimediamarktyou can take this terminal with 256 GB of storage by 823.65 euros. Google Pixel 9 Pro XL 128GB * Some price may have changed from the last review A mobile with excellent value for money This Google Pixel 9 Pro XL stands out for having a screen Super acts OLED of 6.73 inches and resolution of 1,344 x 2,992 pixels. Its soda rate is 120 Hz, it reaches a maximum brightness of 3,000 nits and is compatible with HDR10+. The brain of this terminal is the processor Google Tensioner G4accompanied by the titan m2 backproofing. The RAM is 16 GB and its battery has a capacity of 5,060 mAh and admits fast charge at 37 w. Another of the main identity of this Google mobile is its photographic system, formed by a Triple 50+48+48 MP rear camerawhich is capable of recording video at 8k/30fps. On the other hand, the front camera is 42 MP and in the connectivity section, this terminal also stands out: it comes with Wifi 75G, Bluetooth 5.4 and NFC. It works under the Android 15 operating system and you have the Updates to Android 21. Other high -end mobiles that might interest you Realme Smartphone GT 7 Pro 5g mobile phone 12+256gb * Some price may have changed from the last review Xiaomi 15 Ultra – 16+512GB smartphone * Some price may have changed from the last review Some of the links of this article are affiliated and can report a benefit to Xataka. In case of non -availability, offers may vary. Images | Iván Linares (Xataka) and Google In Xataka | The best mobiles, we have tried them and here are their analysis In Xataka | The best price quality price. Your analysis and videos are here

Goal has been spying on everything we did when navigating our Android mobiles. They have caught it in fraganti

by

Billions of users have been exposed to a new system of mass tracking and espionage that both goal and Yandex have used. Both companies have used a technique that took advantage of native applications such as Facebook or Instagram and then monitor everything we did with our mobile browsers. It is a drop more than one glass already Absolutely full: The attack on Our privacy. What happened. A group of researchers from the IMDEA Networks agency and the Universiteit Radboud of the Netherlands, led by teachers Gunes Care and Nerseo Vallina-Rodríguez, published yesterday An extensive technical article. In it they revealed the technique that they have baptized as ‘local Mess’ (‘Lo Local’) and what a goal has been using since September 2024 and Yandex much earlier, since 2017. What does ‘Local Mess’. We will focus on goal and its applications, although the method is analogous in Yandex. Android users who use applications such as Facebook or Instagram could be exposed, because those applications “listened to” what happened in the Browsers installed In our mobile phones using local ports (hence the ‘Local Mess’) with the aim of tracking and monitoring everything we did in our browsers. The user did not know anything. The method allowed these applications to achieve Receive metadata, cookies and commands that were executed in the browsers. The Meta JavaScript code, called Meta Pixel, was loaded silently and without warning as a kind of complement to mobile browsers, and connected to apps such as Facebook or Instagram. Discouraging users. As these researchers explain, the method allowed access device identifiers that are used for advertising systems, called Android Advertising ID (AAID), and that made it possible to associate everything the user did to a real identity (a Facebook or Instagram profile). The result: what we did in the browser was no longer anonymous or private. Neither unknown, nor delete cookies nor anything. This “web-top” method eludes systems that theoretically should protect this type of tracking. Thus, neither Delete cookiesnor navigate in Incognito mode It worked when trying to escape that tracking. In fact, the method “opens the door for potentially malicious applications to fit the user’s web activity,” they explain in the document. Exploiting our “localhost”. Meta and Yandex scripts are slightly different, but both make improper use of unauthorized access to sockets from our localhostthe reserved name that our device has on the local network and that is always the IP 127.0.0.1 (in IPV4) address. The Android operating system allows any application installed with Internet permission to open a listening socket in the Loopback interface (127.0.0.1). The browsers that are executed on the same device also access this interface without the consent of the user or the mediation of the platform. This allows a JavaScript code embedded on web pages to communicate with native Android applications and share identifiers and navigation activity. Millions of affected websites. For the goal to work in the method, he took advantage of his cookie _FBPvery widespread on websites that make use of this platform. According to Builtwith, a website that allows monitoring the adoption of different technologies, that goal Pixel is embedded in more than 5.8 million websites among which are Xataka. There is a search engine in the final part of the study that allows you to know if a website was exposed and the activity in it could be registered by these scripts. Scheme of the attack of the attack. Source: Local Mess. Theoretically, only on Android. The researchers reveal that they only managed to obtain empirical evidence of this technique in Android mobiles. They have not observed such a problem In iOS browsers Or in the applications they evaluated, although they point out that technically achieving something like this on the iPhones is feasible. Browsers protect themselves. Those responsible for the discovery have followed a communication policy responsible for vulnerabilities and have contacted several browse developers. Chrome already has prepared the patchFirefox’s is in development – but it seems not exposed to the problem in the case of goal – Duckdugo has already solved it and Brave was not affected When using a lock list and when required explicit permission of user for communications with the premises. There is no information on the progress of the patch in Microsoft Edge, which was affected. And goal has deactivated that option without further ado. Although browsers have taken measures, they are late. Not because there is no solution or the code has been modified, but because goal has decided to stop using it without saying anything. Yesterday the Meta Pixel script stopped sending packages or making requests to localhost. The code responsible for sending that cookie _fbp, indicate in an update in this report, has been almost completely deleted. Why did this goal? There is a hypothesis about the implementation of this technique by the goal: it could be due to how Google was intended to get rid of third -party cookies in Chrome At some point in 2024. That would have affected companies as a goal, which perhaps would have reacted trying to collect that information with this technique to have a plan B if cookies disappeared. What says goal. In Xataka we have contacted those responsible for the goal in Spain, and we will update this information if we receive an answer. In The Register they indicate that after contacting the company, Meta has indicated that: “We are in conversations with Google to solve a possible communication error in relation to the application of their policies. When we are aware of the concerns, we have decided to pause the function while we work with Google to solve the problem.” What Yandex says. The Russian company has made comments about the problem. Speaking to Android Authority, they explain the following: “Yandex strictly meets the data protection standards and does not discourage user data. The function in question does not collect any sensitive information and its only objective is to improve the customization of our applications. After examining the problems … Read more

I had no idea how to execute a local AI and I just did it in five minutes with my Android. It has been a disaster

by

The career of whom he manufactures The most capable smartphone It will not be alone in its components: AI has become an essential absolute To win this battle. Gemini He is the absolute protagonist, both in Android and in an iOS in which The presence of AI is insignificant. A complete and capable assistant, with the requirement of being permanently connected to the Internet. Google, Gemini’s mother, has decided Give 100% local models a chance in Android. Including those of its competitors. I, who had never executed a local AI on a PC, I have been able to achieve it on my phone in five minutes. 10 Google applications that could have triumphed Installing Google ai Edge Gallery Google has released an open source app so that any user can interact with multimodal AI models. It is completely free, has no publicity, and its only limitation is that it is not published in Google Play Store. You have to Download it from github. It is as easy as clicking on the link and downloading the file, which weighs 115 MB. Once you have downloaded it, click on it and install like any other app. Depending on the customization layer you use and the browser from which you download it, it is possible that the system asks for the occasional permission. Give it without fear, since it is a clean Google app. Once installed, the app will be in the application drawer or desktop of your phone, depending on your layer. Just open it and download the available models to start using it. Using the local AI in my Android This application allows you to use four predetermined models of the local AI: GEMMA-3N-E2B-IT-Int4 GEMMA-3N-E4B-IT-IN4 GEMMA3-1B-IT-Q4 Qwen2.5-1.5b-Instruct Q8 Each of these names is very likely to sound to Chinese, but there is a very simple summary. Gemma are the Google models, and the number behind “E2B, E4B, 1b”, refers to the parameter number of each model (2 billion, 4,000 million, 1,000 million). This means that, in order, Gemma 3 1b is the fastest, but more basic model, followed by E2B and with E4B to the head as the most complete. On the Qwen side, it is a Chinese model developed by Alibaba, with 1.5 billion parameters, and quite logical precision. These are the models that the app brings, but we can install models on our own. What I win and what I lose using the local The Google app is designed to execute three scenarios of use: Images questions: problem solving, identifying objects, etc. Prompt Lab for summaries, rewriting texts and code generation. Ai Chat: Multi -urnal conversations with AI The local AI is somewhat more limited in these scenarios of use, but in return it is more private. It runs 100% on your phone, without connection to servers. This minimizes response latency, something that I wanted to verify with a fairly simple test: the translation of a text not too extensive. “If the computing power on each chip continue to grow exponentially, Moore real Demanded Were Useful in Business Applications, Too. Defense Contractors Thought About Chips Mostly As a Product that Could Replace Older Electronics in All The Military Systems. “WHEN US DEFENSE SECRETARY ROBERT MCNAMARA REFORMED MILITARY PROCUREMENT TO CUT COSTS IN THE EARLY 1960S, CAUSING WHAT SUB IN THE ELECTRONICS INDUSTRY CALLED THE” MCNAMARA DEPRESSION, “FAIRCHILD’S VISION OF CHIPS FOR CIVILS SEEMED PRISON Off-The-Shelf Integrated Circuits for Civilian Customers Even Sold Products Below Manufacturing Cost, Hoping to convince more customers to trym. “ Chatgpt (4th): 5 seconds Deepseek: 19 seconds The local Google (Gemma 3 1b): has not understood the instruction. Google local (Gemma 3 E2b): 26 seconds. Google local (Gemma 3 E4b): 34 seconds. Google local (Qwen 2.5): 16 seconds. The results are quite irregularand that is that the app is in its initial stage. The main problem I have found is that sometimes it is difficult for him to execute the model, even in cases of light models such as Qwen. In a Samsung Galaxy S25 Ultra I had to restart the app on some occasion, since I did not execute the prompt. I have also wanted to prove how simple problems solve with the visual understanding of images, using the heavy model E4B of Gemma. It has also been quite disaster. In the first attempt I have asked him to solve all the problems. He has succeeded. In the second, I have asked him to solve only the first. He was wrong and, after telling him that it was wrong, he has err. Yes, he does a good job recognizing elements in the images, but it still has a hard time going further (Google just lets his Gemma models choose for these tasks). The conversational assistant works well, with its limitations, but well. It serves to talk and help us with any task that we can think of that you do not need a real -time search on the Internet. If you are especially jealous with your privacy, and want to execute the without any server connection, it is a relatively valid alternative. Google has this app as a test laboratory. At least for now it is not a real alternative, much less, to Gemini. Image | Xataka In Xataka | Artificial Intelligence on your PC: The best free tools to install models of AI as Deepseek, call, mistral, gemma and more

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.