Summary: Threads head Connor Hayes previewed a redesigned web interface that adds direct messages, a navigation sidebar with shortcuts to saved posts and insights, and a cleaner single-feed layout replacing the current multi-column design. DMs, which launched on mobile in June 2025, will roll out on web “over the coming weeks,” bringing one-on-one chats, group conversations of up to 50, and media sharing to the platform’s most engaged desktop users as Threads surpasses 450 million monthly active users and begins scaling its global advertising business.

Threads is getting a redesigned web interface that adds direct messages, a navigation sidebar, and quicker access to features that were previously buried in the mobile-first layout. Connor Hayes, who took over as head of Threads in September 2025, previewed the changes in a post on the platform this week, writing that “web is an important part of how our most engaged users interact with Threads, and we’ll be investing more here going forward.” Messages on the web version are not yet publicly testing, Hayes said, but users should “start to see them appear over the coming weeks.”

The redesign replaces the current multi-column layout with a cleaner single-feed view anchored by a left-side navigation rail. The sidebar includes shortcuts to saved posts, performance insights, activity, notifications, and the ability to switch between feeds, all features that exist on the mobile app but required multiple taps or profile navigation to find on the web. The result looks significantly more like X’s desktop layout, which is either a pragmatic design choice or an admission that the format Threads was trying to replace turned out to be the right one.

DMs finally reach the desktop

Direct messages launched on the Threads mobile app in June 2025, nearly two years after the platform itself launched. The web version has operated without them since, meaning that the users Hayes describes as “most engaged,” those who use Threads on a computer, have been unable to access one of the platform’s core communication features. The web rollout will bring one-on-one chats, group conversations of up to 50 people, emoji reactions, and the ability to send photos, GIFs, and stickers.

Threads has been building out its messaging infrastructure steadily. In January, it launched a basketball mini-game within DMs. In February, it began testing a shortcut that converts the phrase “DM me” in a post into a clickable link that opens a direct message. The messaging system is built on Instagram’s infrastructure, which gives it reliability but also ties it to a platform with different privacy expectations and content norms.

The 💜 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

The redesign preview came one day after Hayes showed changes to how replies look on mobile. Replies under a post will now be indented to make conversation threads easier to follow, a feature rolling out on iOS and currently testing on Android.

The competitive context

Threads has grown faster than any social platform in history and now has more than 450 million monthly active users, with daily active users estimated at roughly 137 to 141 million. In January, Similarweb data showed Threads had surpassed X in daily mobile users, 141.5 million to 125 million, a milestone that would have seemed improbable when the app launched as a text-based companion to Instagram in July 2023.

The growth has come alongside a broader decline of X under Elon Musk’s ownership, which has pushed users, advertisers, and publishers toward alternatives. Bluesky, which raised $100 million in its Series B and has grown to 43 million users under new CEO Toni Schneider, has captured a vocal segment of the market. But Threads’ integration with Instagram’s 2 billion-plus user base gives it a distribution advantage that no standalone competitor can match.

The web redesign is part of a shift from growth to retention. Threads has the users. What it has lacked is the feature depth that makes a platform indispensable for the power users who drive conversation and content creation. DMs, a proper desktop experience, and improved reply threading address the specific complaints that have kept some users treating Threads as a secondary platform rather than a primary one.

Monetisation and Meta’s broader bet

Meta began rolling out ads on Threads globally in late January 2026, after testing in the US and Japan throughout 2025. The rollout uses Meta’s existing Ads Manager and supports image, video, and carousel formats through both Advantage+ and manual campaigns. Early pricing has been lower than Facebook and Instagram, with CPMs estimated at $3 to $8 and cost per click at $0.30 to $1.50, reflecting the early stage of advertiser competition on the platform. Evercore ISI analysts have projected Threads advertising revenue of $8 billion by the end of 2025 and $11.3 billion by 2026.

The advertising rollout gives the web redesign commercial significance beyond user experience. Desktop users tend to have higher engagement times and are more valuable to advertisers. A web interface that keeps users on the platform longer and adds messaging, which increases session frequency, directly supports the revenue trajectory that analysts are projecting.

Hayes was appointed to lead Threads in July 2025, taking over from Adam Mosseri, who had been running the platform directly alongside Instagram. Hayes previously served as Meta’s VP of product for generative AI and spent 14 years at the company in various product roles, including a stint growing Instagram Reels. Mosseri said at the time that “given Threads’ maturity, we think we need a dedicated app lead who can focus all of their time on helping Threads move forward.” The web redesign and DM rollout are the most visible results of that dedicated focus.

Threads is also the largest platform running on the ActivityPub protocol, allowing users to share posts to Mastodon, WordPress, and other fediverse-compatible services. Meta says it has interacted with over 75% of all fediverse servers, though full account portability is not yet available.

The redesign is incremental rather than transformative. It brings the web version closer to feature parity with the mobile app, which is itself still catching up to the feature set that X has built over 17 years. But for a platform that has Meta’s resources behind it, 450 million monthly users in front of it, and a growing creator economy to support, the gap between what Threads offers and what its most engaged users expect is closing faster than most new platforms manage. Hayes is signalling that the web is where the next phase of that closure will happen.

Filling a gap that fans of its retro-inspired speaker range have long identified, Wharfedale has introduced the Heritage Centre.

This new speaker is a dedicated centre channel speaker that’s been built to integrate with Wharfedale’s Linton, Super Linton, Denton, and Dovedale models that have made the Heritage Series one of its most successful lines in recent memory.

The absence of a centre speaker has been a barrier for Heritage owners wanting to build a multichannel home cinema system as the range has until now been limited to stereo pairs.

That’s left buyers to either mix in a mismatched centre channel or go without one entirely when configuring a 3.1 or 5.1 channel setup. Now that’s no longer an issue.

Advertisement

Wharfedale’s solution draws directly from the Super Denton’s driver architecture, adopting the same three-way configuration used across the broader Heritage range to keep the technical foundation consistent across the full speaker family.

That configuration pairs twin 165mm woven Kevlar bass drivers with a 50mm fabric dome midrange and a 25mm fabric dome treble unit, with all three driver types adapted directly from those developed for the Super Denton.

The midrange driver covers the 900Hz to 2.7kHz frequency band, the range most responsible for vocal clarity and dialogue intelligibility in film and television. The treble unit uses a damped rear chamber to push its resonant frequency well below the crossover point to keep high-frequency reproduction clean across a wide listening area.

Cabinet construction uses layered particle board and MDF bonded with a resonance-damping adhesive. It’s a build approach designed to distribute panel resonances across multiple frequencies rather than concentrating them at a single audible point. The internal bracing adds further control over cabinet colouration.

Advertisement

Peter Comeau, Wharfedale’s Director of Acoustic Design said: ““The Heritage Series was originally conceived purely for the enjoyment of stereo music, but the speakers’ richly expressive sonic qualities lend themselves perfectly to other forms of AV entertainment. When the demand for a dedicated centre speaker for people building multichannel systems with Linton and Denton speakers became clear, we embarked on the project with the rigorous attention to engineering detail applied to every Heritage model.”

Advertisement

Real-wood veneers in walnut, mahogany, or black oak finish the cabinet to a hand-polished satin lacquer, maintaining visual consistency with the full Heritage range across all three finish options.

The Wharfedale Heritage Centre arrives in late May, priced at £649, available in walnut, mahogany, or black oak to match whichever Heritage speaker system it sits in.

Advertisement

Simple and iconic, IKEA’s Billy bookcase has been around since the 1970s, and over 140 million have bene sold worldwide. The classic wood and white finishes are timeless, but now it’s got a new look for 2026 with a limited-edition blue version.

A bold piece of furniture like this needs the right styling, and as TechRadar’s Homes Editor, I like making it pop by teaming it with black and white for a striking effect. This compelling cobalt bookcase would look particularly good in a home office, with an IKEA Kallax desk in black/brown, and white accessories.

Disclaimer: Unless otherwise stated, any opinions expressed below belong solely to the author.

In Mar, Singapore’s Ministry of Manpower (MOM) released its annual summary of job creation efforts that took place in the year before, highlighting the skills and expertise needed in both PMET and non-PMET professions.

For this piece, we focus on PMET roles—Professionals, Managers, Executives and Technicians—where many of Singapore’s best and most sought-after jobs are often concentrated (although some statistics may overlap).

Why are new jobs created in Singapore?

MOM’s analysis begins with the fundamental question: are Singaporean employers looking for replacements or are they genuinely adding new openings to their offer?

Fortunately, last year brought the highest reading yet: 49.3% of vacancies were for completely new positions. This means that local companies are looking for more people and are not simply rotating staff.

What’s more, a record-high share of this expansion is being driven by businesses creating entirely new functions. In 34.7% of cases, job growth came from new roles rather than the expansion of existing operations, which, unsurprisingly, still accounts for the majority at 55.8%.

It suggests that 2025, despite the fears caused by the US tariffs, was a very dynamic year, and companies still ventured into new areas.

Where are the jobs created?

Where are those new areas found, then?

Well, as has been typical over the past few years, the industry with the highest share of fresh openings remains Information & Communications, where close to three-quarters of vacancies are for roles that did not exist before.

It is followed by Construction (though it’s most likely driven by non-PMET employment), as well as Professional Services and Finance & Insurance, where more than half of the jobs on offer are new.

That is great news, of course, given that some of the best-paid roles are found in Singapore’s corporate sector.

Who are these jobs for?

Qualified people, naturally, but as we explained on Vulcan Post recently, paper degrees matter less and less, even for PMETs, where 70% of employers stated that academic qualifications are not their main consideration.

This doesn’t mean they don’t matter at all, but if all you have is a paper rather than practical experience, your job search may be considerably longer, just as it is a problem for employers to recruit workers for some vacancies for more than six months (listed in the table below).

Lack of skills and experience are the two primary reasons they remain in the market, with not enough talent available to fill them. And it’s not like employer expectations are huge, but over half of those looking for PMET specialists expect at least two to five years spent on the job somewhere before.

Only one in five is willing to employ complete newbies.

Here’s a more specific breakdown by industry:

If you’re a fresh graduate or someone without experience at a particular job, your best chance may be to look for something in the public sector, as it is the most open to candidates without a long CV. It also pays well and looks for applicants with greater educational attainment.

So, if you have a degree but are struggling for work, perhaps take a look at what state administration or education are offering.

How much do they pay?

Finally, let’s talk about the money.

Here’s the list of the Top 10 most in-demand PMET jobs, compiled from the data collected in 2025, together with the salaries you can expect.

Top 10 PMET Vacancies in 2025

The podium is occupied by the same jobs as last year, with a switch between second and third places. But it’s the teachers who are still in the highest demand, while the upper pay band places their earnings at over S$100,000 per year. Not bad.

Software developers and related IT experts are still highly needed—and highly paid, as are Electronics Engineers, System Analysts and Financial Advisers.

Other jobs may not be quite as lucrative, but their availability should make up for it, as many Singaporeans (including young grads) are looking for their way into the labour market.

• Read other articles we’ve written on Singaporean businesses here.

Featured Image Credit: Google Street View

In short: Stanford’s 2026 AI Index Report finds the performance gap between the best American and Chinese AI models has collapsed to 2.7%, down from 17.5-31.6 percentage points in May 2023, despite the US spending 23 times more on private AI investment ($285.9 billion vs $12.4 billion). China leads in AI patents (69.7% of global filings), publications (23.2% of global output), industrial robot installations (9x the US rate), and energy infrastructure, while AI talent migration to the US has dropped 89% since 2017.

The performance gap between the best American and Chinese AI models has collapsed to 2.7%, according to the 2026 AI Index Report published this week by Stanford University’s Institute for Human-Centered Artificial Intelligence. In May 2023, the gap was between 17.5 and 31.6 percentage points across major benchmarks. As of March 2026, Anthropic’s Claude Opus 4.6 leads the global leaderboard with an Arena score of 1,503, while ByteDance’s Dola-Seed-2.0-Preview sits at 1,464, a difference of 39 points. DeepSeek’s R1 reasoning model briefly matched the top US model in February 2025, and American and Chinese models have traded the lead multiple times since.

The 423-page report, the most comprehensive annual assessment of the global AI landscape, documents a situation in which the United States spends 23 times more on private AI investment than China but leads on the only metric that arguably matters, model performance, by less than three percentage points. The question the report raises without quite answering is whether that spending advantage is sustaining American leadership or whether China has found a way to compete without it.

Where each country leads

The United States dominates private AI investment, with $285.9 billion in 2025 compared with China’s $12.4 billion. California alone accounted for $218 billion, more than 75% of the US total. American companies produced 50 notable AI models last year, compared with China’s 30, though China’s count doubled from 15 the previous year while America’s grew more modestly. The US hosts 5,427 data centres, more than ten times any other country.

China leads in volume. Chinese researchers produced 23.2% of all global AI publications and 20.6% of citations, compared with 12.6% for the US. Chinese entities filed 69.7% of all AI patents worldwide. China installed 295,000 industrial robots in the most recent reporting period, nearly nine times the 34,200 installed in the United States. And China’s electricity reserve margin has never dipped below 80%, twice the necessary capacity, while the US power grid suffers from decades of underinvestment that the report identifies as a potential bottleneck for AI infrastructure growth.

The investment figures come with a significant caveat. The report notes that private investment data “likely understates” China’s actual AI spending because the Chinese government channels resources through guidance funds and state-initiated investment vehicles that do not appear in private capital databases. The 23-to-1 spending ratio may be less dramatic than it appears.

The talent crisis

The most striking finding may be about people rather than models. The number of AI scholars moving to the United States has dropped 89% since 2017, with 80% of that decline occurring in the last year alone. The report describes the fall as “precipitous.” Switzerland now ranks first in the world for AI researchers and developers per capita.

The talent migration data complicates the narrative that American AI leadership is secure because of its investment advantage. If the researchers who build frontier models are increasingly choosing not to come to the US, the spending premium buys hardware and infrastructure but not the intellectual capital that turns compute into capability. DeepSeek demonstrated in January 2025 that a Chinese lab could match Silicon Valley’s best with a fraction of the resources. The talent data suggests the conditions that produced DeepSeek are strengthening, not weakening.

What AI can and cannot do

The report documents performance gains that would have seemed implausible two years ago. On SWE-bench, a coding benchmark, model performance rose from 60% to near 100% in a single year. On graduate-level science questions, model accuracy hit 93%, above the expert human validator baseline of 81.2%. Google’s Gemini Deep Think won a gold medal at the International Mathematical Olympiad. On Humanity’s Last Exam, a benchmark designed to be unsolvable, frontier models gained 30 percentage points in a year.

But the report also documents what it calls a “jagged frontier.” The top model reads analog clocks correctly only 50.1% of the time. Robotic manipulation systems achieve 89.4% success in simulation but only 12% in real household tasks. Nearly half of the 500-plus clinical AI studies reviewed used exam-style questions rather than real patient data, and only 5% used actual clinical records. The gap between benchmark performance and real-world reliability remains wide in domains where errors have consequences.

Adoption, trust, and regulation

Generative AI reached 53% population adoption within three years of launch, faster than the personal computer or the internet. Eighty-eight per cent of organisations report using AI. Four in five university students now use generative AI tools. But the US ranks 24th globally in adoption at just 28.3%, behind Singapore at 61% and the UAE at 54%.

Public trust is lower still. Only 31% of Americans trust their government to regulate AI, the lowest figure of any country surveyed and well below the global average of 54%. The expert-public disconnect is a central theme of the report: 73% of AI experts expect a positive impact on jobs, compared with 23% of the general public. Only a third of Americans expect AI to make their jobs better.

Forty-seven countries now have active AI legislation, but only 12 have enforcement mechanisms. Documented enforcement actions rose from 43 in 2024 to 156 in 2025. Compliance costs vary eightfold between jurisdictions. The EU AI Act entered full enforcement in January 2026, but the broader regulatory picture is one of fragmentation rather than coordination.

The environmental cost

Training xAI’s Grok 4 produced 72,816 tonnes of CO2 equivalent, roughly the emissions of driving 17,000 cars for a year. AI data centre power capacity reached 29.6 gigawatts globally, enough to power New York State at peak demand. The environmental section of the report reads as a counterweight to the performance gains: the models are getting better, but the cost of making them better is scaling alongside the capabilities.

What the numbers mean

The headline finding, that China has nearly closed the performance gap with the US, will dominate the policy conversation. But the report’s deeper implication is about the relationship between spending and outcomes. The United States invested $285.9 billion in private AI capital last year. China invested $12.4 billion. The performance gap between their best models is 2.7%. Meanwhile, AI talent migration to the US has collapsed, China dominates patents and publications, and Chinese infrastructure investment in energy and manufacturing dwarfs America’s.

The open-versus-closed source debate adds another dimension. The top closed model now leads the top open model by 3.3%, up from 0.5% in August 2024, and six of the top ten Arena models are closed-source. The performance advantage of proprietary systems is widening, which favours the American companies that dominate the closed-source tier but also means the open-source models that have driven China’s catch-up may face diminishing returns.

Employment data for software developers aged 22 to 25 fell nearly 20% since 2022. One-third of surveyed organisations expect AI to reduce their workforce in the coming year. The Foundation Model Transparency Index dropped from 58 to 40, with most frontier models reporting nothing on fairness, security, or human agency. Documented AI incidents rose 55% in a year.

The Stanford AI Index does not make policy recommendations. It presents data. But the data in the 2026 edition tells a story that should unsettle anyone who assumes American AI dominance is durable. The US leads on investment and model performance. China leads on talent pipeline, patents, publications, robotics, and energy infrastructure. The performance gap is 2.7% and shrinking. The spending gap is 23 to 1 and growing. One of those trends is sustainable. The report leaves it to the reader to decide which one.

• Most organizations cannot reliably track returns from email campaigns
• Strong ROI exists only for companies that actively measure performance
• Many teams rely on content generation without deeper optimization strategies

Email marketing continues to generate strong returns, yet many organizations still lack clarity on whether those returns are actually being realized.

The recent Sinch Mailgun’s Email Impact Report 2026 analyzed insights from more than 400 billion emails sent in 2025 and surveyed over 1,200 email senders, finding fewer than half of organizations can reliably track return on investment from their email programs.

The Apple Watch Series 11 is a smartwatch worth upgrading to. It’s the best smartwatch for iPhone owners, and the base price is reasonable. It tends to swing back and forth in cost between its MSRP of $399 and a sale price of $299. Right now, it’s back to a match of that low price, meaning it’s the perfect time to make the upgrade if you’ve been hunting for a new Apple Watch.

Note that this sale price is for the 42-millimeter case size without GPS. If you want cellular connectivity or the larger 46-millimeter case, you’ll pay a bit more. But across all retailer options, nearly every color-and-size combination is discounted. Available finishes include Gold, Natural, and Slate titanium options, and Rose Gold, Silver, Space Gray, and Jet Black if you opt for aluminum.

The Apple Watch Series 11 finally has a battery that can last at least a full day. An actual full day, as in 24 hours, meaning you can wear it while you’re at the gym and while you’re sleeping. This will allow you to better take advantage of its myriad of tracking capabilities. (As the owner of an Apple Watch Series 8, I often consider upgrading for this reason alone.) Aside from the typical fitness stats and workout tracking, plus the AI-enabled Workout Buddy feature, this watch can monitor for signs of hypertension and track blood oxygen levels. It also has Fall Detection and satellite messaging capabilities (on models with cellular connectivity).

All in all, while new tech is neat, it’s not always worth upgrading for. But last year’s Apple Watch introduces meaningful changes that you’ll notice in your day-to-day life. If you’re still rocking an older model, or you’re shopping for your first smartwatch, this is absolutely worth considering—especially at this sale price. Afterward, check out our favorite Apple Watch bands to spruce up your new gadget.

Vornado Box Fan Model 80X for $100: While most people who need a box fan are, frankly, going to run out to Walmart or Home Depot and grab one for 20 bucks, you should be aware that there exists a Rolls-Royce of box fans. “It has 99 speeds,” the brand’s rep told me when it came out. “Yeah, right,” I thought. But, sure enough, this thing actually has 99 speeds, accessible via up and down buttons. I have no idea under what circumstances one might need this many speeds, but there they are. It’s also got a kickstand to reduce wobbling, a digital display, and a 1-to-12-hour timer. Plus, the silver-and-black casing looks good—like you meant to have it in your house, not a remnant from that one summer your AC broke during a heat wave.

Photograph: Kat Merck

Shark TurboBlade (Bladeless) for $250: Though this 2025 blade-less model is billed as a tower fan, it doesn’t look or act like any tower fan I’ve ever seen. It evokes a windmill more than it does a fan, with a horizontal bar that sits on a telescoping base, like a big “T.” The ends of the bar, which are articulated, feature the vents, and each end can be bent straight up, straight down, or at any point in between for fully customizable air direction. The whole bar can also be turned vertically to look more like an “I,” if you’d rather have a tall, thin breeze as opposed to a long, thin breeze. It has all the usual features you’d expect of a fan at this price point, including 10 speeds, oscillation, a magnetic remote, and three settings, including “Sleep,” which makes sense as the TurboBlade, in its “T” configuration, is about the right height for a bed. It’s a great choice if you need airflow in different directions at once, but be forewarned that it makes a fairly loud, jet engine-like whine, which is noticeable even on lower settings. There’s also now a TurboBlade Heat + Cool ($400), which adds a 1,400-watt heater to the middle, but WIRED reviewer Matthew Korfhage tested it and didn’t find the heat feature to be worth the extra $150.

Shark FlexBreeze for $200: This was my favorite misting fan of last year. I love that it’s rechargeable, so it can be used without an electrical outlet nearby, and I love that the head detaches from the pedestal with legs that fold out, allowing it to double as an easy-to-transport floor fan. Shark claims the FlexBreeze can reduce nearby ambient temperature by 10 degrees with the misting attachment. Though I was never able to measure a reduction of more than 6 degrees using multiple thermometers, the difference in air temperature using the FlexBreeze versus without is dramatic enough to make the difference between an unbearable summer dinner outside and a pleasant one. However, the mist deployed by the detachable misting attachment (Shark now makes a version with a tank, but I haven’t tried it) is a bit on the heavy side—it made most of my deck quite wet and dampened the clothes of anyone sitting within 5 or so feet. On the plus side, this meant the mist didn’t immediately blow away, as was the case with the FlexBreeze’s portable sibling, the HydroGo (below).

Photograph: Kat Merck

Shark FlexBreeze HydroGo for $150: I loved the original Shark FlexBreeze (above), but not the fact that it had to be connected to a hose, so I was very excited to see a rechargeable, portable version in fun colors. Shark says it can run for 30 minutes with the mister consistently on, or 60 minutes in “interval mode,” and after testing it at my son’s soccer practices, I found these estimates to be more or less accurate. However, the mist that comes out of the middle is so fine and in such a small stream that it blew away quickly before it had a chance to cool anyone, unless they were sitting just inches from it.

Lasko Whirlwind Orbital Pedestal Fan for $85: This fan looks a lot like Dreo’s TurboPoly 508S, and indeed sports some of the same features—it oscillates vertically 105 degrees or horizontally 150 degrees, it’s quiet (I clocked 27 dB on low), and it’s got a remote. It’s not smart, it doesn’t have RBG lights, and there are some occasional noises from the oscillation, but if you’re looking for a more affordable pedestal fan that offers 3D oscillation, this honestly isn’t a bad option.

Among the global brands at AXPONA 2026, Mimic Audio did not have the biggest booth or the loudest presence, but it ended up being one of the more worthwhile stops in the EarGear section. The Chicago dealer, owned by TJ Cook, was positioned between Campfire Audio and Austrian Audio and only a few steps from the always swamped ZMF booth, which made it easy to overlook in the rush. That would have been a mistake. Mimic first caught my attention before the show when it supplied the AudioByte components for the Von Schweikert pre-event, paired with NUR Audio’s Harmonia.

My initial listen there was promising, but with the Von Schweikert VR.thrity or Ultra 7 commanding the room and the Harmonia’s open-back design letting all of that noise pour in, it was impossible to draw more than a few early conclusions. That made a return visit at AXPONA essential, where I sat down with all three NUR models on display for a longer listen and a better sense of what this Italian headphone brand is actually bringing to the table.

NUR Audio Headphones: Italian Design, Planar Magnetic Ambitions

NUR Audio is not some legacy brand trading on decades of goodwill. It was founded just northeast of Rome by Angelo De Mattia and feels very much like a passion project finding its footing in a crowded category. Right now, the Harmonia open back is the only model you can actually buy, priced at $3,750, while the Shanti open-back reference and Miah closed back are still listed as coming soon with pricing to be determined. That split matters because NUR is already drawing a line between audiences. The Harmonia is built for listening at home, while the Shanti and Miah mark the start of a professional series aimed at engineers who need precision more than romance.

The two open-back designs share a lot of DNA. Similar materials, similar construction, and very similar planar magnetic drivers. The Miah goes a different route with a dynamic driver inside a closed back design, which should make it the more practical option for studio work or less than ideal environments. All three, however, are physically imposing. Think Audeze LCD-4 sized ear cups and the kind of weight that can turn a long session into a short one if the ergonomics are off. Early impressions suggest NUR understands the problem. The suspension system is well padded, the clamp feels reasonable, and the weight distribution does not immediately raise red flags.

The real test, as always, will be whether that comfort holds up after a few hours rather than a few tracks.

Using the AudioByte stack (more on that soon), I was able to spend time with all three NUR models and come away with a clearer sense of how each is voiced. With both the Shanti and Miah still in prototype form, nothing here should be considered final, but the direction is already apparent.

The NUR Harmonia is a large-format open-back planar magnetic headphone built around a 105mm PEEK diaphragm and a double-sided toroidal magnet system using high-grade N52 neodymium magnets. That combination is designed to deliver fast transient response, low distortion, and wide bandwidth, which is reflected in the rated 8Hz to 55kHz frequency response.

With a 48 ohm impedance and 107 dB/mW sensitivity, it should be relatively easy to drive for a planar of this size, though it will still benefit from a capable amplifier. The dual 3.5mm cup connections allow for balanced operation out of the box, with either 4.4mm or XLR cables included, along with a 6.35mm adapter for single-ended use. At 630 grams, it is firmly in the heavyweight category, making the suspension system and overall ergonomics critical for longer listening sessions.

The Harmonia leans toward a clean, controlled presentation with a touch of warmth that you don’t always get from planar magnetic designs. Bass has solid presence without sounding pushed, the midrange comes across as slightly lush with very good detail retrieval, and the treble extends well past what my ears are willing to admit at this point. It strikes a balance that feels intentional rather than trying to impress on first listen.

The Shanti prototype shifts gears toward a more analytical presentation. It is crisper, more forward in its detail, and less forgiving overall. The name was a bit of a clue, but the tuning confirms it. This feels like the model aimed squarely at those who want to dissect recordings rather than relax into them.

The Miah, as the closed-back option, moves in a different direction. It is warmer and a bit thicker sounding than the two open-back models, which is not surprising given the design. Detail is still present across most of the range, but the top end has slightly less extension and sparkle. That trade-off is typical for closed-back headphones, especially ones that appear to be targeting studio use rather than chasing an artificially boosted sense of air.

The Bottom Line

I came away impressed enough to spend a good amount of time talking with TJ Cook about getting all three NUR models in for proper review once they hit the market. That says more than any quick show impression. AXPONA has no shortage of big names pulling crowds, and it is easy to fall into the trap of chasing logos instead of sound. The problem is that you end up walking right past booths like Mimic Audio and missing some of the more interesting listens of the weekend.

The NUR lineup, paired with the AudioByte components, proved to be far more than a curiosity. It was one of those setups that rewarded anyone willing to sit down, block out the noise, and actually listen. Not perfect, not finished in two cases, but clearly headed somewhere worth paying attention to.

Expect a deeper dive once review samples land. In the meantime, NUR Audio is a brand to keep on your radar, and if you happen to be in the Chicago area, Mimic Audio is absolutely worth a visit.

Where to buy: $3,750 at Mimic Audio

Related Reading: