The AirPods 4 are the best entry-level AirPods yet. | Photo by Chris Welch / The Verge It’s a good day for Apple fans on a budget. Not only did the company introduce a new budget phone, its entry-level AirPods 4 are also on sale. Right now, you can snag a pair of Apple’s AirPods 4 starting at just $89 ($30 off) at Walmart. The model with active noise-cancellation are also down to their second best price to date, going for $119 ($60 off) at Amazon and Walmart.AirPods 4Where to Buy: $129 $89 at Walmart $179 $119 at Amazon (with ANC) $179 $119 at Walmart (with ANC)The AirPods 4 are the best and most comfortable entry-level earbuds from Apple yet. They deliver fuller bass along with clearer, more detailed sound than previous models, and even now match the AirPods Pro in terms of voice call quality. They also boast an upgraded IP54 rating for water and dust resistance, meaning they can withstand light rain and sweaty workouts. As with other Apple headphones, they also play nicely with Apple devices, offering conveniences like automatic switching, access to hands-free Siri, and Apple TV support.If you’re willing to spend $30 extra, the ANC model offers all the same features while also doing a surprisingly good job of tuning out the world. They can’t rival the AirPods Pro, but they still dramatically cut down on ambient and low-frequency noise on planes and in offices, particularly with music playing. The ANC earbuds also offer a transparency mode that sounds natural, along with support for tracking via Apple’s Find My App, a wireless charging case, and live translation through an iOS 26 update.

Anthropic is making it easier to switch to its Claude AI from other chatbots with an update that brings Claude's memory feature to users on the free plan, along with a new prompt and dedicated tool for importing data from other chatbots. These upgrades could allow users who have been using rivals like OpenAI's ChatGPT or Google's Gemini to quickly copy the data their preferred AI has collected on them and bring it over to Anthropic's chatbot. That way, they don't have to "start over" teaching Claude the context and history their previous chatbot already knows. The option to import and export memories from Claude has been available since O …Read the full story at The Verge.

Microsoft has a little bit of an embarrassing situation on its hands. Its big entry into the handheld market is the ROG Xbox Ally, and its more powerful brother, the Ally X. Both of these consoles are mighty, well-made devices, but they reportedly get even better if you strip Windows off them and replace it with either SteamOS or Bazzite.

Most local AI users assume that a bigger model is always better than smaller models. It is always expected that more parameters equal smarter outputs. I also had similar assumptions. For a long time, 7B models felt like entry-level, and I thought of upgrading to a larger model.

The OrangePi Neo was going to be an exciting launch for fans of both portable gaming and open-source software. It's a console that doesn't look too unlike a Steam Deck, but it runs Manjaro Linux with KDE Plasma 6 and flatpak installations. The predicted price point for this console looked very reasonable: 450$ for the 16 GB RAM version, and $550 for the 32GB model, with both of them sporting 512GB of storage space.

I didn't build a home lab just to run Plex and Pi-hole. I mean, that's some of the reason, but I also did it so that I could form an understanding of how things work. What better way to learn about enterprise-level IT infrastructure than to immerse yourself in it at home, where little mistakes won't lead to the boss breathing down your neck or a barrage of new help desk tickets being filed?

Tax season is here for those who reside in the US, and the deadline to file is coming quickly, set as it is for every year on April 15, 2026. For most people, it's going to be a fairly simple affair, and you'll be able to file for free using the official website. But for others, their finances could be a bit more complex, which may require some added help from third-party software.

To say that AI has reshaped a lot of fields in one way or another would be an understatement. Programming, though, might be the field where the pace of change feels faster than almost anything else. It began with copying snippets of code out of a ChatGPT window and pasting them into an IDE. Then came AI-powered inline suggestions and tab completion. Now, we're in the era of agentic coding tools like Cursor, Antigravity, and Claude Code, where, in addition to generating lines and lines of code, the AI is capable of actually doing tasks for you instead of telling you how to do them. This includes reading your codebase, creating new files, running commands, and building features end-to-end while you simply prompt and watch.

So, you finally dropped $500–$1,000 on a fancy new graphics card after finding a remotely reasonable "deal" and launched your favorite AAA titles to take your new GPU for a spin. An hour went by, but you missed the part where the new GPU was supposed to feel like an upgrade. The games felt the same as they did on your older card, and you were left wondering what the point was. If this sounds familiar, then you probably upgraded your GPU at the wrong time, or at least without thinking about the rest of your build. Multiple reasons can make a GPU upgrade feel underwhelming, from poor raw performance and diminishing returns to system bottlenecks and game selection. Let's get into each of them and unpack what you can do to remedy the situation.

If you're someone who multitasks between two devices for your daily work, being efficient is extremely critical. While you can easily plug in and swap any accessories between two devices, it makes it a lot easier just to have one USB switch that can cater to both. This not only keeps things in order, but it also cuts down on downtime as well.

For most people, it's not very exciting to talk about workstation GPUs. Many gamers may not even know that Nvidia, AMD, and Intel manufacture GPUs not meant for gaming. However, advanced users, home labbers, and professionals know the importance of enterprise GPUs, even the pre-owned models selling on eBay. These graphics cards are barely relevant for gaming compared to consumer GPUs, but even the old models still pack a punch for professional workloads and home servers. If you're after a capable yet economical GPU for your home server, a used workstation card may be way better than a consumer card, even at the same price. Enterprise GPUs are built differently and offer unique advantages that are missing even on modern consumer cards.

A Raspberry Pi makes a rough desktop the moment you ask it to behave like a modern PC. The browser feels heavier than it should, even before you start piling on extensions and tabs. Storage hiccups show up in little stutters that are easy to ignore on a faster machine, but hard to unsee once you notice them. Even simple updates can take long enough to make you wonder what else you could have done with that time.