AMD has launched its first open-source large language models (LLMs) under the OLMo brand, aiming to strengthen its position in the competitive AI landscape led by giants like Nvidia, Intel, and Qualcomm.
AMD OLMo is a series of 1-billion parameter large language models trained from scratch using trillions of tokens on a cluster of AMD Instinct MI250 GPUs. They are designed to excel in reasoning, instruction-following, and chat while embracing an open-source ethos that allows developers access to data, weights, training recipes, and code.
“Continuing AMD tradition of open-sourcing models and code to help the community advance together, we are excited to release our first series of fully open 1 billion parameter language models, AMD OLMo,” AMD said in a statement.
AMD’s open-source approach positions OLMo as an accessible and scalable option for companies seeking alternatives in AI technology. The model can be deployed in data centers or on AMD Ryzen AI PCs equipped with neural processing units (NPUs), allowing developers to leverage advanced AI directly on personal devices, the statement added.
“AMD is following Nvidia’s lead by expanding into the large language model (LLM) space alongside its well-established strength in computing hardware — a direction that Intel and Qualcomm have not yet fully embraced,” said Abhigyan Malik, practice director at Everest Group. “By fostering an open ecosystem, AMD enables developers to innovate and build diverse applications through a network effect.”
According to Malik, this strategy amplifies AMD’s core value proposition, particularly in driving demand for its underlying hardware, including AMD Instinct MI250 GPUs and Ryzen CPUs, where “AMD seeks to create lasting market impact.”
Extensive training and fine-tuning
The OLMo series follows a detailed three-phase training and fine-tuning process, according to AMD.
Initially, OLMo 1B was pre-trained on a subset of the Dolma v1.7 dataset using a transformer model focused on next-token prediction. This helped the model grasp general language patterns. In its second phase, the OLMo 1B was supervised and fine-tuned (SFT) on multiple datasets to refine its capabilities in science, coding, and mathematics.
The final model, OLMo 1B SFT DPO, was optimized with Direct Preference Optimization (DPO) based on human feedback, resulting in a model that effectively aligns its responses with typical user expectations.
Competitive performance and benchmark success
In internal benchmarks, AMD’s OLMo models performed well against similarly sized open-source models, such as TinyLlama-1.1B and OpenELM-1_1B, in multi-task and general reasoning tests, the company claimed. Specifically, its performance increased by over 15% on tasks in GSM8k, a substantial gain attributed to AMD’s multi-phase supervised fine-tuning and Direct Preference Optimization (DPO). ‘
In multi-turn chat tests, AMD claimed, OLMo showed a 3.41% edge in AlpacaEval 2 Win Rate and a 0.97% gain in MT-Bench over its closest open-source competitors.
However, when looking at the broader LLM landscape, Nvidia’s GH200 Grace Hopper Superchip and H100 GPU remain leaders in LLM processing, particularly for large, multi-faceted AI workloads. Nvidia’s focus on innovations like C2C link, which accelerates data transfer between its CPU and GPU, gives it an edge, providing a speed advantage for high-demand inference tasks such as recommendation systems.
Intel, while slightly behind in peak speed, leverages its Habana Gaudi2 accelerator for cost-effective yet robust performance, with future upgrades planned for increased precision. ‘
Meanwhile, Qualcomm’s Cloud AI100 emphasizes power efficiency, meeting the needs of organizations seeking high AI performance without the extensive energy demands associated with Nvidia’s high-end systems.
AMD’s OLMo models also showed strong performance on responsible AI benchmarks, such as ToxiGen (for toxic language detection), crows_pairs (bias assessment), and TruthfulQA-mc2 (accuracy). These scores reflect AMD’s commitment to ethical AI, an essential focus as AI integration scales across industries.
AMD’s position in the AI market
With its first open-source LLM series, AMD is positioned to make significant inroads in the AI industry, offering a compelling balance of capability, openness, and versatility to compete in a market currently led by Nvidia, Intel, and Qualcomm.
However, AMD’s ability to close the gap will depend on how well its open-source initiative and hardware enhancements keep pace with rivals’ advances in performance, efficiency, and specialized AI capabilities.
“AMD’s entry into the open-source LLM space strengthens the ecosystem, potentially lowering the operational costs associated with adopting generative AI,” said Suseel Menon, practice director at Everest Group.
AMD’s move into LLMs places it against established players like Nvidia, Intel, and Qualcomm, who have gained market prominence with their proprietary models.
“This move also puts pressure on proprietary LLMs to continually innovate and justify their pricing structures,” Menon added.
Analysts believe AMD’s unique open-source strategy and accessibility aim to attract enterprises and developers looking for flexible, affordable AI solutions without proprietary constraints.
“For large enterprises with long-term data privacy concerns, AMD’s open-source model offers a compelling alternative as they navigate AI integration,” Menon added. “By building a cohesive, full-stack AI offering that spans hardware, LLMs, and ecosystem tools, AMD is positioning itself with a distinct competitive edge among leading silicon vendors.”
Cloud architects, data security engineers, and ethical hackers are among the highest-paying skills that can be attained through IT certifications — and AI technology didn’t even make the list.
Online learning platform Skillsoft analyzed the top reported salaries of IT professionals around the world to find the highest-paying certifications and developed a list of more than 20.
This year’s list shows that cloud computing skills remain in high demand and can be quite lucrative. The AWS Certified Security Specialty training jumped from sixth-highest to the top-paying certification this year to now command a $204,000 annual salary on average — a up 22% or $40,000 over last year.
The presence of certifications for Google Cloud Platform (GCP), AWS, Azure, and Nutanix also highlights the value of a diverse cloud skillset, as organizations adopt multi-cloud or hybrid cloud strategies, according to Skillsoft.
Its list is similar to one published earlier this year by job search platform Indeed, which also placed an AWS certification in the No. 1 slot. (Indeed found AWS Certified Solutions Architects could earn from $133,200 to $246,900 a year at some firms.)
“So, are they worth it? For those looking for any of the above, it’s a resounding yes,” Skillsoft said a blog post. “But, earning a certification takes time, effort, and often money.”
Are certifications worth the price?
Earning a certification led to pay raises, promotions and new jobs, according to Skillsoft. In addition to AWS training, rounding out the top five certifications were:
Google Cloud – Professional Cloud Architect, averages $190,204.
Nutanix Certified Professional – Multicloud Infrastructure (NCP-MCI) v6.5, averages $175,409.
Gartner Research, in an August report, also found that AWS Certified Cloud Practitioners and Microsoft Certified Azure Fundamentals certifications were top upskilling opportunities for tech workers. Other IT certifications with fast-growing demand this year are in cybersecurity, including the CISSP certification, CISA, and CompTIA Security+, according to Gartner. (The latter — IT certifications from the Computing Technology Industry Association (CompTIA), a non-profit trade association — were also among the general class of top certifications on multiple lists.)
“While learning new technology skills is vital, the ability for employees to demonstrate practical expertise through industry-recognized certifications is increasingly valued,” Gartner said. “Though they may not be a mandatory prerequisite for every position, certifications can empower individuals and organizations alike.”
“Our data suggests that tech professionals skilled in cloud computing, security, data privacy, and risk management, as well as able to handle complex, multi-faceted IT environments, will be well positioned for success,” said Greg Fuller, vice president of online learning platform Codecademy Enterprise. “Overall, the IT job market is characterized by a significant imbalance between supply and demand, which continues to drive salaries higher.”
What’s happening with AI training?
While AI certifications have not yet to the top of IT certification lists, the increasing emphasis on data privacy and compliance is closely tied to the rollout of AI technologies. And while AI skills are gaining popularity, it often takes time for certifications to gain traction, Fuller said.
“Right now, what we see with areas like AWS Security at the top is that organizations are still preparing for large scale AI rollouts,” he said. “So more adjacent skills are on this year’s list. Ultimately, it’s a mix of certifications being a bit slower to evolve and adjacent skills rising in criticality.
“In the meantime, the backbone of AI is cloud, so getting cloud certified is a good first step. Then, look at some of the more specialized Cloud AI certifications,” Fuller added.
Recruitment and talent consulting firm WilsonHCG released a report this week indicating that while AI certifications might not be on the top 20 lists, there is rising demand for AI skills across sectors. The market for AI-skilled workers is expanding, too, with 5,898 average monthly job postings in October, according to WilsonHCG.
The rise in the number AI-focused certifications reflects a significant increase from the 12-month average of 5,147, driven by heightened interest in roles like data scientist, AI research engineer, and machine learning engineer.
Companies such as TikTok, Apple, Google, Amazon, and Deloitte are among the most active in AI recruitment, underscoring the technology’s growing adoption in sectors from tech to finance and professional services, according to WilsonHCG.
“The need for AI skills extends beyond traditional tech positions. Companies are seeking professionals across a range of roles, including Founding AI Engineer and Senior Software Engineer for AI products,” WilsonHCG said in its report. “This trend is reshaping hiring practices and job titles as more organizations prioritize data-driven and AI-enabled functions across departments.”
Skills continue to matter more than formal education
Skills-based hiring approaches that emphasize strong work backgrounds, certifications, assessments, and endorsements, continue to dominate the tech industry. And soft skills are becoming a key focus of hiring managers, even over hard skills.
Elise Smith, co-founder and CEOofPraxis Labs, an AI-based learning platform, said she has worked with enterprises like Google, Uber, and ServiceNow to help senior leaders develop the skillsets needed for “new-age talent retention” and collaboration in the workplace.
“As workplaces continue to transform — whether its emerging technologies like genAI transforming how we work or sociopolitical conflicts that cause disruption to our workflows — human skills will become more and more important,” Smith said.
What’s often missing from higher education is a focus on skills building around interpersonal communication, conflict resolution, critical reasoning, and the ability to determine fact from opinion or misinformation. “What once may have been called soft skills will be seen as power skills, and workforces who focus and develop these skills will differentiate in market outcomes,” Smith said.
While building relations and moving beyond “transactional trust” in the workplace can be challenging — especially for a hybrid global workforce — it’s important to build skills around workplace connection.
“When managers are skilled in asking open-ended questions, coaching disengaged team members, learning more about individuals’ backstories and contexts, and encouraging them in their work, teams thrive,” she said. “These are the skillsets we help our clients and their people leaders develop.”
The UK government has introduced an AI assurance platform, offering British businesses a centralized resource for guidance on identifying and managing potential risks associated with AI, as part of efforts to build trust in AI systems.
About 524 companies now make up the UK’s AI sector, supporting more than 12,000 jobs and generating over $1.3 billion in revenue, the UK government said. Official projections estimate the market could grow to $8.4 billion by 2035.
The Mozilla Foundation, the nonprofit organization behind the Firefox open-source browser, said it has laid off about 30% of its employees as part of a reorganization to increase its “agility.”
As of 2023, the foundation had between 80 and 300 employees, according to varying reports. A spokesperson declined to say how many employees the company has now.
Established in 2003, the group is best known for its development of the Firefox web browser, as well as its advocacy for internet privacy, digital rights, and freely-available, open-source software.
A Mozilla Foundation spokesman said the non-profit is reorganizing teams to boost agility and impact as it accelerates efforts for “a more open and equitable technical future. That unfortunately means ending some of the work we have historically pursued and eliminating associated roles to bring more focus going forward,” Brandon Borrman, vice president of Mozilla’s communications, said in a statement to Computerworld.
The non-profit arm is distinct from the Mozilla Corporation, which is the for-profit company responsible for generating revenue through products like the web browser. The corporation employs a much larger number of people, likely 700 or more.
The Mozilla Foundation’s executive director, Nabiha Syed, said in an email last week that two of the foundation’s major divisions — advocacy and global programs — are “no longer a part of our structure,” according to a TechCrunch report.
Contrary to reports, however, Borrman said the restructuring will not impact its goal of open-source and free internet advocacy. “On the contrary, advocacy is still a central tenet of Mozilla Foundation’s work,” he said. “Fighting for a free and open internet will always be core to our mission, and advocacy continues to be a critical tool in that work. We are in the process of revisiting our approach to it.”
Along with the Mozilla Foundation, Mozilla currently consists of five organizations: the Mozilla Corporation, which leads consumer product-based work; Mozilla Ventures, a “tech-for-good” investment fund; Mozilla.ai, an AI R&D lab; and MZLA, which makes Thunderbird.
In 2020, the Mozilla Corporation cut about 25% of its 1,000-person global workforce, saying that the coronavirus pandemic’s impact on economies “significantly impacted our revenue.”
Borrman said the layoffs did not affect any of the other Mozilla entities.
Let’s face it: For many people, web browser performance could well be more important than general PC performance.
Browser makers are wising up to this, too. Google Chrome just introduced new performance controls, while Microsoft Edge has attempted to stand out with its own browser performance options. And every web browser out there has long fought over the title of fastest in the land.
So let’s talk browser performance — and how you can get more of it, specifically when working within Windows. In a world where websites feel like they’re getting heavier and heavier, upgraded browser performance means everything from faster load times and a better all-around browsing experience to more reliable all-around PC performance and longer laptop battery life.
Want more Windows PC tips? My free Windows Intelligence newsletter delivers all the best Windows tips straight to your inbox. Plus, you’ll get free in-depth Windows Field Guides as a special welcome bonus.
Windows web browsing boost #1: Cull your extensions
Does your browser feel inexplicably slow? Before you do anything else, I’d recommend pruning any installed browser extensions. Add-ons can be useful, but they can also add some serious overhead to your browsing. They may be always running in the background, or they may run some code on each web page you load.
In Google Chrome, you can click the main three-dot menu icon > Extensions > Manage Extensions to see a list of what’s installed. From there, you can disable or remove them. Other browsers have a similar menu and mechanism, potentially with slightly different placement and phrasing.
You might want to try disabling a few browser extensions first to see if your browser feels faster. If not, you can easily re-activate them by flipping their switches back on in that same area of your browser’s settings.
Windows web browsing boost #2: Put those tabs to sleep (or keep them awake)
Modern web browsers — including Chrome, Edge, and Mozilla Firefox — all have features that put tabs to “sleep.” If you don’t use a tab for a while, your browser will stop it from running. It won’t be able to use resources in the background. When you click back to the tab, your browser will reactivate it.
This saves memory, and it also stops pages in background tabs from using CPU resources. Overall, it will boost your browsing speed.
However, in some cases, it could slow things down. Perhaps you often find that you switch back to a tab and your web browser quickly reloads it. If that’s a problem, you’ll want to make your browser stop putting tabs to sleep — especially if you have a powerful computer with a lot of RAM and a fast CPU. (You can also tell your browser to stop putting specific websites to sleep if it causes a problem with a website.)
To control tab suspending:
In Google Chrome, click menu > Settings and select “Performance” in the left pane. Look under “Memory Saver” and choose an option: Moderate, Balanced, or Maximum. You can also disable Memory Saver entirely — or add websites you never want Chrome to suspend to the “Always keep these sites active” list there.
In Microsoft Edge, click menu > Settings and select “System and performance” in the left pane. Use the “Save resources with sleeping tabs,” “Put inactive tabs to sleep after the specified amount of time,” and “Never put these sites to sleep” options to control this behavior.
In Mozilla Firefox, this feature is always activated — unless you dig deep into Firefox’s settings to turn off tab unloading.
Modern browsers can suspend, sleep, or unload tabs to save system resources. These are all names for the same trick.
Chris Hoffman, IDG
Windows web browsing boost #3: Preload more pages
Your web browser of choice can “preload” some pages. In other words, it might load them in the background if it thinks you’ll visit them. If you do, the page loads very quickly — because by the time you’re looking at it, it’s already loaded in the background and ready to go!
Most browsers offer different preloading options, some of which are more aggressive than others. And preloading has some potential privacy implications, as your browser might load links you wouldn’t have clicked. But, for maximum speed, you’ll probably want the most aggressive preloading options available.
To control preloading:
In Google Chrome, click menu > Settings and select “Performance” in the left pane. Scroll down to the “Preload pages” option. For maximum speed, ensure “Preload pages” is active and that it’s set to “Extended preloading.”
In Microsoft Edge, click menu > Settings and select “Cookies and site permissions” in the left pane. Click “Manage and delete cookies and site data,” and ensure “Preload pages for faster browsing and searching” is activated.
Windows web browsing boost #4: Check your browser’s task manager
Want to see what’s actually using CPU and memory? Modern Chromium-based web browsers — including Chrome, Edge, Brave, Arc Browser, and more — have task managers that will show you. (Firefox has something similar, too.)
In a Chromium-based browser, just right-click an empty spot on the tab bar and select “Task Manager” or press Shift+Esc to open it.
You will see a list of processes — including open web pages, browser extensions, and browser components — along with how much CPU and memory they’re using. If your web browser is mysteriously slow, this is a good place to check: You might spot an open web page that’s dragging everything down, and you can close it from here. You can also click the “CPU” heading to sort processes by CPU and see the most CPU-hungry items at the top of the list.
In Firefox, you can access something similar by plugging about:processes into Firefox’s address bar and pressing Enter. (The Shift+Esc shortcut will work, too!)
Your browser’s task manager will show you if a web page or browser extension is hogging system resources.
Chris Hoffman, IDG
Windows web browsing boost #5: Clear your browser cache (or stop clearing it)
Ah, the browser cache. As you browse, your web browser remembers the pages you visit and the things you type in a history, it stores images and other bits of downloaded pages in a cache, and it keeps cookies with information from websites — like your sign-in status.
Many people frequently clear this browser cache. If your browser is slow, you can try clearing browsing data. In fact, Microsoft’s official Edge browser documentation says “Clearing your browser data on a regular basis will improve the performance of your browser” — and who am I to argue with Microsoft? Surely, it understands how its own browser works.
Clearing that data is worth a shot. But, conversely, if you’re clearing your browsing data too regularly, you might want to stop doing that. The browser cache is there to speed things up: Your browser can pick images and other bits of web pages out of its cache rather than redownloading them, which improves load times when you revisit a page.
You’ll find options for clearing browsing data in your browser’s menu, but you can also just press Ctrl+Shift+Delete to quickly open the browser-history-clearing tool.
Clearing your browser data can speed things up — but clearing your browser cache too aggressively can also slow down page-load times.
Chris Hoffman, IDG
Windows web browsing boost #6: Scan for malware
We have to talk about malware for a minute. Whenever a PC is running mysteriously slow, malware is always one of the first things you should check for.
Windows web browsing boost #7: Switch up your ad-blocker
When it comes to ad-blockers, one thing people don’t often talk about is the fact that such systems can both speed up and slow down your browsing. The speed-up part is obvious: By refusing to load advertising resources on web pages, ad-blocking plugins reduce download size and produce a lighter page that opens more quickly.
But there’s also a slow-down factor: Ad-blockers might also run extra code on the pages you visit, increasing memory use and making them take longer to load.
Different ad-blockers will have different effects on performance. There’s been a lot of controversy about Google Chrome’s switch to Manifest V3 and how it stops the popular “classic” uBlock Origin ad-blocker from functioning. But here’s the thing: While the new way Chrome blocks ads with Manifest V3 is less powerful, it’s also faster. So if you happen to be using uBlock Origin and install the new uBlock Origin Lite, you could see improved page load speeds.
That’s because those new Manifest V3-compatible ad-blocker extensions work by providing a list of resources they want to block. The Chrome browser engine then blocks those resources. That means the ad-blocking browser extension itself doesn’t have to get involved and run a bunch of code on the pages you access.
If you want a speed boost, it’s something worth chewing over. If you’re not yet using any ad-blocker, consider installing one. If you are using an ad-blocker, consider switching — for example, to something like uBlock Origin Lite.
Just bear in mind that you might occasionally break a page; you might need to turn it off for a page if you run into issues.
Windows web browsing boost #8: Try a fresh browser profile
To be clear: I’m not recommending you run out and factory-reset your PC! But popular browsers have built-in “fresh start” tools that will clean up your browser profile and its settings, wiping away any configuration changes, disabling extensions, and erasing cached files to give you a like-new browser. It’s worth a shot.
Here’s how to do it:
In Google Chrome, click menu > Settings and select “Reset Settings” in the left pane. Use the “Restore settings to their original defaults” option.
In Microsoft Edge, click menu > Settings and select “Reset Settings” in the left pane. Click the “Restore settings to their default values” option.
In Mozilla Firefox, click menu > Help > Troubleshoot Mode. You can then click “Refresh Firefox” in the dialog box that opens.
It’s a good way to start over. And hey — if you’re experiencing any kind of PC performance issue, browser-related or otherwise, the old standby advice is always good: Try turning it off and on again.
There’s more where this came from! My free Windows Intelligence newsletter delivers all the best Windows tips straight to your inbox. Plus, you’ll get free copies of Paul Thurrott’s Windows 11 and Windows 10 Field Guides (a $10 value) just for subscribing.
Zooming out, will Donald Trump’s victory in the US presidential race give Apple more bargaining power when negotiating with European regulators — and to what extent will the ongoing US anti-trust investigation of the company (shaky as it is) gain presidential support?
Those could be the kind of questions Apple CEO Tim Cook is asking himself this morning as the former President inches toward a new administration in 2025. We can surmise this based on what Trump said during the campaign, when he explained how Cook rang him up to complain about the fines levied against the company by Europe.
What Trump told Cook
Speaking on a podcast, the incoming President alleged that he told Cook he would not let the EU “take advantage of our companies.” If he keeps that promise, this suggests we may have a new entrant in the Europe versus Apple (and hence, Big Tech) ring. With more regulatory investigation — including the first-ever potentially $38 billion fine under the Digital Markets Act (DMA) — headed Apple’s way in Europe, could the former and future president intervene some how?
It’s hard to tell; after all, most people have become cynical about politicians and the promises they make (and later break) on the campaign trail. We don’t know yet whether the next Trump administration will keep promises made on the way to the White House, or just cherry pick those it wants to keep and ignore the rest.
If the new government does choose to support American tech business against what Republicans might see as overreach by the EU “deep state,” then the next time Europe decides to take a few billion from Cupertino things might not go quite so easy.
When the gloves are off, what happens?
While it is understandable that Europe desperately wants to blunt foreign behemoths in the tech sector in a strategic attempt to support the growth of its own players in that space, it is possible that plan may fail. After all, as events in Valencia, Spain, suggest, Europe has other problems.
The thing is, given the inherent nativism of so much of Project Trump, can European regulators afford to play hardball here? Future history will tell. But there is no doubt the answer to these questions does matter to many in the US tech sector, and also, inevitably, to supply chain partners elsewhere.
One thing that does seem likely is that Apple’s investment in manufacturing in India will continue to accelerate, as the new administration seems set to continue the policies toward China it maintained last time it held power. Cook’s strategic vision to set up shop in India seems likely to pay long-term dividends, as does the considerable work the company has already done and continues to do to repatriate jobs to the US — an ongoing effort on which it has spent hundreds of billions of dollars so far.
Cook, meanwhile, will continue to follow his own approach toward engaging with others who hold opinions he perhaps does not share. “Personally, I’ve never found being on the sideline a successful place to be,” he told employees in 2019.
That approach led him to become one of Trump’s top tech advisors during the first administration.
Cook’s way of doing things also seems to have won some support from Trump, who recently said he thought that if Cook didn’t run Apple it wouldn’t be nearly as successful as it is now. “I think Tim Cook’s done an amazing job,” he said. “And I’m not knocking Steve Jobs.”
Trump also seemed impressed at the eye-watering size of Europe’s fines levied against Apple, which he characterized as “a lot.” With all of this in mind, it is perhaps important to note that Trump in 2019 said Cook has a direct line to the (now) newly-re-elected President.
The art of the deal
Might this contribute to the art of some kind of new EU deal? We don’t know that, either, but as America — and the world — digests the election results, it might yet prove an important moment for Apple’s business, too. European regulators need to think about it.
Sometimes a Windows installation simply goes off the rails. Menus don’t open properly, icons start moving around the desktop, File Explorer acts up, apps get weird, and so forth. Enough things can go wrong, or turn strange, that it’s important to understand various basic Windows repair strategies.
Over the past decade, one of the chief techniques in my own repair arsenal for Windows 10 and 11 has been what’s sometimes called an “in-place upgrade install” or an “upgrade repair install.” (Spoiler alert! This is absurdly easy to do in Windows 11 version 23H2 and later.)
Before going into the details of how to perform such a maneuver, let’s start with a definition and some explanation.
What is an in-place upgrade install?
An in-place upgrade install involves using the Windows OS installer to replace all the operating system files for Windows 10 or 11 on a PC. Basically, you’re using the installer’s setup.exe program to reinstall the same OS back over itself. This leaves user files entirely alone, retains many settings and preferences and, best of all, leaves already-installed apps and applications unchanged. It does, however, overwrite operating system files more or less completely. And in so doing, it often repairs a balky or misbehaving OS and returns it to normal, working condition.
It can take as little as 15 minutes to perform an in-place upgrade install. This maneuver doesn’t require much post-installation cleanup, tweaking, or follow-up activity, either.
Sounds too good to be true — what’s the catch?
Indeed, an in-place upgrade install can provide a quick and effective fix for many, many Windows problems and issues. I use this technique regularly, particularly when I notice that a system is starting to misbehave yet proves resistant to basic repair techniques, such as running the system file checker (SFC) or using the deployment image servicing and management (DISM) image cleanup capabilities.
But an in-place upgrade install is not a universal panacea, and it doesn’t work to cure all Windows ills, either. It is particularly powerless to reverse changes to the Windows registry. (That’s one reason why Microsoft advises users to steer clear or take a registry snapshot they can restore later if they must go where angels fear to tread.)
Here are some additional key considerations that determine the suitability of an in-place upgrade install for a Windows installation.
Requirements for an in-place upgrade install
You must be logged into an administrative account to perform an in-place upgrade install.
Windows 10 or 11 must be running (and keep running) so that you can run the setup.exe installer from inside Windows itself. You cannot run an in-place upgrade install using a bootable Windows installer or when Windows is booted into Safe Mode. (This is what’s called a “clean install,” and if you take this route, it does not save apps, applications, or any settings or preferences.)
You will need at least 9GB plus whatever disk space Windows is using on the drive where it’s running to perform an in-place upgrade install. That’s because the installer renames the running version to Windows.old and lays down a whole new Windows folder for the upgrade it copies to disk. That extra ~9GB or so is needed for work space during the install process.
The Windows installer you use must be the same edition (Home, Pro, Education, or Enterprise), the same language (for example, en-US for United States English, en-GB for British English), the same “bittedness” (32- or 64-bit for Windows 10; Windows 11 is 64-bit only), and the same build (or newer) as the Windows image it upgrades and repairs.
Post-installation checklist
Please note further that some cleanup or customization may be required once the in-place upgrade install has completed. You should check all these things, some of which may require some additional time and effort to complete:
Custom fonts and customized system icons will be absent following an in-place upgrade install. If you want them back, you’ll have to restore them manually.
Wi-Fi connections may need to be re-established (including providing SSIDs and passwords). Occasionally, networks may change from Private to Public and will have to be reset properly.
Windows Update will only be current as of the date of the image file used for the in-place upgrade install. All subsequent updates must then be applied from Windows Update to make the new installation completely current.
By default, Windows turns System Protection off. After an in-place upgrade install, System Protection must be turned on to enable capture and use of restore points if you want them.
The previous installation’s OS files in the Windows.old folder consume substantial disk space. Once things are working properly, run Disk Cleanup as Administrator to clear out those old files and recover the up to 35GB of disk space they typically consume. (You can see that folder represented in the WizTree disk space analyzer in Figure 1.)
Figure 1: On a Windows 11 test PC, the Windows.old folder (outlined in white at top center) comes in at 34.7GB after an in-place repair install.
Ed Tittel / IDG
Once you’ve chewed through this list and pondered all the potential gotchas, performing an in-place upgrade install is easy — ridiculously so in Windows 11 and straightforward in Windows 10 with the right Windows ISO in hand.
Windows 11: Click one button to perform an in-place upgrade install
For versions 23H2 or later, you need only navigate to Settings > System > Recovery, then click the Reinstall now button next to “Fix problems using Windows Update,” as shown in Figure 2. It uses Windows Update for whatever version of Windows 11 you’ve got installed, matching the base version and installed updates exactly. It’s literally a one-button upgrade option.
Figure 2: Click “Reinstall now” and Windows Update does the rest.
Ed Tittel / IDG
That said, this does take some time: I’ve had it take anywhere from 35 to 110 minutes to complete on various test machines. (Newer PCs with fast CPUs finished faster than older PCs with slower ones.) For that reason, Windows 11 users might wish to use the speedier ISO installation method described next, which also works for Windows 11 versions prior to 23H2 and for Windows 10.
Windows 10 or 11: Perform an in-place upgrade install from an ISO or USB media
An ISO, also called an “ISO image,” is a large single file that originally represented the contents of an entire optical disk — a CD, DVD, or Blu-ray Disc. This format is well-suited for installing a large, complex operating system such as Windows because it can bundle up all the programs, files, configuration data and so forth that go into installing such an operating system on a PC.
Windows 10 users can visit Microsoft’s Download Windows 10 page to grab its Media Creation Tool (MCT) for Windows 10. Running the MCT offers an option to build a Windows 10 ISO file. This approach works only for current versions of Windows 10, though. If you need something older (or newer, like a Windows Insider ISO) you may want to turn to HeiDoc.net’s Windows ISO Downloader or UUP dump instead. If you want to grab an ISO for Windows 11, you can head to Microsoft’s Download Windows 11 page or UUP dump.
Remember: the ISO you use to perform the repair install must match the version you’re trying to repair. Your running OS can tell you everything you need to know to pick an ISO for an in-place upgrade repair install. See Windows 10 Forums and Windows 11 Forum for details on how to elicit that info.
Once you’ve got the right ISO, you’ll need to do a little prep work before beginning the in-place upgrade process:
Be sure to log in to Windows with an administrative account.
If Windows runs on a drive that’s encrypted, you’ll need to suspend or turn off encryption before performing the in-place upgrade install. After the install completes, you can turn it back on again.
If the target PC runs UEFI (the Unified Extensible Firmware Interface), turn off fast boot and secure boot before starting the in-place upgrade install. Again, you can turn it back on after it’s done.
Disable or uninstall any third-party antivirus or security software that may be running (anything other than Windows Defender, in other words). Once again, you can reinstall or reenable it once the install is complete.
With that out of the way, running the repair install is dead simple:
Mount the ISO.
Navigate to the root of the virtual “CD Drive” into which the ISO’s contents get loaded.
Run the setup.exe file.
If you’ve got a bootable USB medium (normally a flash drive), you can skip step 1. Open the drive in File Explorer and run setup.exe.
When the Windows installer gets going, accept the license terms, select Upgrade this PC now, allow updates, and click Next. Windows grabs updates, switches over to the installer OS image, and gets itself ready to run. You must then accept the license terms and allow the OS to start the actual in-place upgrade.
By default, the installer keeps all personal files and apps on the target machine. This is what you want, so there’s no need to dig into the “Change what to keep” item on the “Ready to install” page. Just be sure that both “Install Windows 10” (or “Install Windows 11”) and “Keep personal files and apps” are checked on that screen.
As the in-place upgrade runs, the circular progress indicator shows that it’s upgrading Windows, from 1% to 100%. Figure 3 shows the corresponding Windows 10 screencaps; Windows 11 screens are similar but not identical.
Figure 3: The Windows 10 installer prepares to resinstall the OS (top) and then grinds through the initial installation phase prior to the first reboot (bottom).
Ed Tittel / IDG
After that completes, the installer takes you through some additional setup screens where you have the option to customize settings or take the express route to completion. Once that is complete, you’ll sit through a number of colored screens as the installer puts the finishing touches on your in-place Windows 10 or 11 upgrade.
For the vast majority of PCs, it will take less than 20 minutes for this process to complete. Older, slower PCs may take half an hour or more, but that has not been my experience. This means that when an ISO is available, this process goes much faster than the simpler one-button Windows 11 option described in the previous section. I tend to prefer this for my Windows 11 repairs for that reason and because I keep current ISOs around as a matter of habit and choice.
After the installation
Please remember to check the list of items in need of possible attention and effort when the install is finished, as outlined earlier in the story.
By default, Windows keeps the Windows.old folder around for 10 days after such an install. If you’re sure things are working, you can remove it sooner. On the test Windows 11 PC whose Windows.old folder weighed in at just under 35GB in Figure 1, running Disk Cleanup (cleanmgr.exe) returned nearly 25GB worth of disk space, as you can see in Figure 4. That process took about 5 minutes to complete.
Figure 4: Disk Cleanup offers to recover 24.4GB on the repaired Windows 11 PC.
Ed Tittel / IDG
A strategy for using Windows in-place upgrade installs
Knowing that I can perform an in-place upgrade install quickly and easily has really changed my outlook on Windows troubleshooting. Except for hardware problems (or driver issues, which tie directly into hardware as well), if I find myself spending half an hour troubleshooting a Windows problem, I’m already asking, “Is it time for an in-place upgrade install?” Once that time spent stretches past one hour, there has to be a compelling reason why it’s not a good idea to perform an in-place upgrade install to keep me laboring away at other things.
Simply put, an in-place upgrade install is a great solution for resolving trying or opaque issues with Windows — as long as the target OS is still running well and long enough to run setup.exe through the first of the three or four reboots typical during Windows 10 or 11 installation. If you can make it to the first reboot, the new OS takes over after that anyway, and most problems will be fixed.
Over the past year, I’ve either experienced directly or read about an in-place upgrade install fixing a lengthy laundry list of vexing problems, including these:
Issues with system fonts, icons, thumbnails, and other presentation matters.
Networking problems with Wi-Fi and Ethernet, and with misbehaving or absent network interfaces.
Start and program menu issues, navigation and taskbar problems, and application window issues related to placement and sizing.
Flaky or erratic behavior from File Explorer, Edge, and UWP apps. Ditto for general OS instability (slow performance, unreliable system utilities, or frequent OS errors).
Otherwise intractable Windows Update issues (Windows 10 or 11 PCs can’t or won’t download or install updates or feature upgrades).
These days, if a Windows 10 or 11 problem proves hard to diagnose or fix, I’ll turn to an in-place upgrade install as a next or inevitable step in the troubleshooting and repair process. Much of the time, it provides the fix that’s needed. Savvy admins and power users could do worse than give it a try. Cheers!
This article was originally published in March 2018 and updated in November 2024.
At this point, I think it’s safe to say the current wave of “AI everything” isn’t all it’s cracked up to be.
Let’s not sugarcoat it: The vast majority of generative-AI gobbledegook is unreliable, impractical, and more about marketing than any manner of real-world benefit for us — the (alleged) humans meant to be benefiting from it. But here and there amidst all the overhyped hullaballoo, a genuinely useful possibility pops up and peeks its way through the metaphorical curtain.
For some folks, that’s absolutely been the case with the AI-powered summarizing options appearing in apps and services across the tech universe these days. One click, tap, or spoken request — and boom: In a matter of moments, your favorite neighborhood AI genie starts summing up mountains of text and distilling endless-seeming info down into a succinct summary for you.
And if there’s one place where such a feat could be especially intriguing, it’s in our ever-overloaded Android notification panels.
With Apple already embracing the idea in its recently rolled out Apple Intelligence suite (for better or, erm, maybe for worse, in that scenario) and Samsung now supposedly considering something similar for its Android-based Galaxy gadgets, I’ve heard more than a few questions from my fellow Android-appreciating animals about if and how it’d be possible to have notifications automatically summarized here in the land o’ Android.
My friend, I’ve said it before, and I’ll say it again: On Android, where there’s a will, there’s a way.
Lemme show ya the secret.
[Psst: Grant yourself even more noteworthy notification powers with my new Android Notification Power-Pack — six smart enhancements that’ll change how you use your phone.]
How to summarize notifications on Android — no waiting required
If you want your Android device’s notifications summarized as they stream in — without any waiting for Samsung, Google, or anyone else to make such a feature officially available — a clever little app called Sum Up AI Notification Summaryis the little-known key to making it happen.
And yes, ironically, the app’s name could do with some summarization — or at the very least a visit from the Department of Redundancy Department. But that teensy bit of amusement aside, it really is an incredible tool.
And it’s pretty easy to get going on whatever Android device you’re using, no matter who made it or how old it might be.
In fact, you’ve got just three key steps to consider:
Android notification summary step 1: Install and connect
First things first, install the Sum Up app from the Play Store (obviously — right?!). It’ll cost you a whopping $1.65, up front and in a single one-time payment. And it doesn’t require any disconcerting permissions or collect any form of personal data, as per its Play Store privacy policy.
Now, before we go any further: The app uses Gemini to handle the heavy lifting, and you have to connect it to your Google account before it can access that resource. It sounds complicated on the surface, but I promise you: It really isn’t at all difficult to do — provided you’re in an area where access to Gemini is generally available.
Here’s all there is to it:
Tap the box at the top of Sum Up’s main screen to get a Google Gemini API key.
On the screen that comes up next, tap the button labeled “Create API key,” then tap the search box and select “Gemini API” (or any other option, really — it doesn’t actually matter!) in the list that shows up.
Tap “Create API key in existing project,” then tap the button to copy the key you’ve created.
Now, head back to the Sum Up app. Tap the “Gemini API key” line there, paste the key you just copied into that field, and tap “OK” to save it.
See? Told ya it wasn’t bad.
And Google won’t charge you a single penny for this privilege, either, within any normal-use scenarios — going all the way up to 15 requests per minute and 1,500 requests per day. (If you’re dealing with more notifications than that, you’ve got bigger fish to fry!)
Android notification summary step 2: Authorize and configure
Now that that part’s out of the way, you’ve just got a handful of relatively simple settings to make your way through before sitting back and enjoying your newfound Android notification summarizing powers.
Ready?
First, within that same main Sum Up setup screen, tap “Grant notification access,” then activate the toggle to allow the app to read and interact with your notifications (for what I hope are obvious reasons).
Head back to the setup screen and tap “Allow posting notifications,” then tap “Allow” on the prompt that pops up (again, for reasons that should be pretty apparent).
See the slider labeled “Messages threshold”? I’d slide that up to either two or three, to start. Otherwise, the system will summarize notifications even when there’s only a single message involved — and if you ask me, that isn’t really needed or helpful in any way.
And last but not least, flip the toggle next to “Retain original notification actions” into the on and active position. That’ll allow options like replying directly to a message or marking something as read to be present and available even on your summarized notifications.
Setting up Sum Up to summarize your Android phone’s notifications is much easier than it looks at first glance.
JR Raphael, IDG
And that’s pretty much it! If you want, you can look through the “Per-app settings” option to see which specific apps Sum Up will watch and then summarize notifications from for you. It works with most email and messaging-oriented apps, including Slack, Google Messages, WhatsApp, and Gmail — but if there’s any particular app where you don’t want it to summarize, you can easily uncheck it in that area.
Sum Up lets you select exactly which apps you do — and don’t — want to see notifications summarized for.
JR Raphael, IDG
Take a deep breath and treat yourself to a quick crumpet break, if you’re feeling peckish. We’re almost done here — and your Android notification summary machine is basically now up and running.
Android notification summary step 3: Watch and adjust
At this point, all that’s left is to wait and watch as Sum Up works its magic.
I’d suggest keeping a close eye on the summaries for a few days and seeing what seems helpful vs. more annoying than valuable to you. You might find, for instance, that when it comes to regular texts, you prefer seeing every individual message from Google Messages by itself — but you might appreciate the summaries when it comes to notification-heavy work-related Slack conversations.
Each of these notifications took three incoming Slack messages and summarized them into a single skimmable sentence.
JR Raphael, IDG
Whatever you decide, the power to make it happen is now firmly in your hands. And that, m’dear, is the power of Android — summed up as succinctly as can be.
Extending the standard refresh cycle of laptops in a corporate environment can significantly reduce a workplace’s carbon footprint, according to new workplace sustainability research. However, challenges in ensuring the older laptops still meet Windows OS requirements, and other cost issues, could offset these environmental gains, experts said.
The Atos Sustainable Workplace report, published this week and aimed at providing insights to help improve corporate social responsibility, found that by refreshing laptops every four years rather than three, organizations achieve up to a 25% reduction in carbon and related emissions without downgrading device performance or user experience.
Moreover, by using what it called “data-driven, condition-based device refresh combined with remanufacturing,” organizations can even extend the life of laptops within an organization for eight to 10 years, according to the report.
Most (79%) of a laptop’s carbon footprint is produced during manufacturing, with each new device creating roughly 338kg of carbon dioxide equivalent (CO2e) before use, according to Atos. This means that extending the lifecycle of corporate laptops can have a huge impact on how sustainable a corporate environment is.
Atos research also found that 76% of large organizations’ laptops can be remanufactured — that is, stripped down and updated with newer parts where needed, and then sent back into the workforce — with the remaining 24% of devices available for refurbishment or recycling to contribute to the circular economy.
Moreover, the study found that employees are more than willing to use their older laptops longer for the sake of environmental benefits, with 75% of them saying they would be willing to keep them longer if it meant a more sustainable workplace.
Real-world issues to consider
Improving sustainability in a workplace is certainly a noble and, in the current global environment, even necessary goal. However, the vision Atos presents for extending laptop lifecycles to achieve environmental gains may not make practical, economic, or even cultural sense for current corporate environments, analysts noted.
One issue with the idea of keeping older laptops for longer is that Microsoft’s support for Windows 10 ends in less than a year. Most organizations are likely in the middle of a migration to Windows 11, which is challenging, as many older PCs may not meet the minimum requirements for the new operating system.
If organizations choose not to migrate, they may soon have to start paying for extended Windows 10 support, which starts at $61 per device in the first year, doubling to $122 per Windows 10 device in year two, and $244 per device for the third and final year.
Indeed, organizations are currently facing an important choice as the end of Windows 10 support looms: how do they want to approach either migration to Windows 11, or staying with older Windows 10 devices, noted Everest Group senior analyst Prabhneet Kaur.
“While component upgrades, like increasing RAM, can help meet Windows requirements, or informal workarounds may bypass them, maintaining robust security and a quality user experience will be tricky,” she said.
‘A lot of work for IT departments’
According recent Neowin numbers, overall share of Windows 10 users is currently at 60.95%, while the overall share of Windows 11 users is at 35.55%. These are consumer numbers; corporate environments often lag behind consumer adoption of new versions of Windows.
If an IT department wants to keep older laptops and migrate them to Windows 11, it will take both time and financial investment to maintain the older devices — neither of which corporate IT departments or budgets historically tend to have much of, another analyst noted.
“TCO [total cost of ownership] increases as the device gets older,” observed Ranjit Atwal, research director in Gartner’s Quantitative Innovation team, adding that remanufacturing laptops “sounds expensive to do every few years.” Overall, the idea of keeping older laptops in corporate circulation for a longer period of time “sounds like a lot of work for IT departments that don’t like change,” he noted.
Sebastien Vibert, solution manager at Atos, said his company’s analysis found that of tens of millions of devices observed, 96% of devices in large organizations are already compatible with Windows 11. This indicates that “a majority of corporate organizations have taken the necessary steps to refresh their devices, ensuring they meet the requirements for Windows 11” and thus support Trusted Platform Module 2.0, a critical hardware component of Windows 11 security.
Even if this is the case, there still remain “many challenges” to Atos’ vision of sustainability through laptop lifecycle extension, Atwal observed.
“There are lots of different parts of the organization to persuade to make the move to this type of sustainable path,” he noted. “However, sustainability is a major factor on the CIO agenda and these considerations are being discussed.”
Smartphone deliveries worldwide increased by 5% during the third quarter of 2024 compared to the same period a year ago, according to a new survey from analysis firm Canalys. That equates to about 310 million units and is the best performing third quarter for the smartphone market since 2021.
Canalys traces the growth to increased demand and an aggressive launch of several new products, including Samsung’s latest smartphone series and Apple’s iPhone 16 series.
During the quarter, Samsung maintained its lead in the market, taking 19% of sales, followed closely by Apple with 18% and Xiaomi with 14%. In fourth and fifth place were Oppo and Vivo, with 9% of the market each.
Canalys expects continued growth for the market through the end of the year.
