Angry Birds to make a full return to mainland China
MIIT says it will expand coverage of the “secondary number refresh” program
Steam now supports completing bundle purchases for friends
Nova Launcher acquired by a new Swedish owner
On January 21, Swedish company Instabridge announced that it has completed the acquisition of Nova Launcher. Instabridge made it clear that Nova Launcher will not be shut down, and that its current priorities are to ensure product stability, maintain compatibility with new versions of Android, and keep up regular maintenance.
However, the company also revealed that it is evaluating the possibility of introducing ads into the free version. Tracking components for Facebook Ads and Google AdMob have already been added to the latest version of the code, and Reddit users have reported seeing ads in the app.
Instabridge said that the paid version, Nova Launcher Prime, will remain ad-free. As for open-sourcing the project, the company noted that responsible open-sourcing involves not only the code itself, but also licensing, security, build systems, contribution workflows, and trademark management, and therefore requires careful consideration. It added that any decision will be publicly communicated once made. Source
Angry Birds to make a full return to mainland China
On January 21, the well-known mobile game Angry Birds announced its return to the mainland China market. Kingsoft Seasun, a subsidiary of Kingsoft Corporation, and Rovio Entertainment today announced the establishment of a strategic partnership, paving the way for the franchise’s full return to mainland China.
As the first step in this comeback, the two companies plan to launch two Angry Birds IP titles. The flagship game, Angry Birds: Classic Returns, will be comprehensively upgraded based on the iconic slingshot gameplay, featuring a 3D engine, dynamic level design, and a brand-new magic system. Angry Birds: Dream Blast will offer a lighthearted bubble-shooter puzzle experience, inviting players to embark on a fantastical journey.
At the same time, to energize the creative community around Angry Birds: Classic Returns, Rovio and Kingsoft Seasun are jointly launching the “Prank Workshop” creator program, inviting community creators to contribute original content, including custom levels and visual art. Source
MIIT says it will expand coverage of the “secondary number refresh” program
At a press conference held by the State Council Information Office on January 21, officials responsible for the “secondary number refresh” initiative said that basic telecommunications operators have already refreshed more than 250 million recycled phone numbers in batches before reissuing them, and have unlinked them from over 1 billion internet applications. The “proactive refresh” service portals in apps and mini programs support removing historical bindings with 239 commonly used internet applications, having processed more than 360 million unbinding requests for over 5.8 million users.
Going forward, the coverage of the “secondary number refresh” program will be further expanded, with efforts to connect more internet applications closely related to everyday life. China Mobile, China Unicom, and China Telecom began rolling out the service in May 2025, allowing users to unbind historical internet accounts with one click. Source
Steam now supports completing bundle purchases for friends
On January 20, third-party Steam data site SteamDB discovered that the “gift to a friend” feature on the Steam platform now supports “bundle completion,” allowing players to help friends complete game bundles they have partially purchased.
When a player gifts a bundle containing multiple games to a friend, the system will automatically detect which titles the friend already owns and adjust the final price accordingly. For example, if a friend already has some of the games included in the bundle, the checkout price will automatically deduct the value of those owned titles. The remaining games that the friend does not yet own will still benefit from the bundle’s combined discount. The goal is to make it easier for players to help friends complete their game collections without paying again for content they already own. Source
RedMagic launches the RedMagic 11 Air and multiple new products
Zhipu releases and open-sources the GLM-4.7-Flash model
Netflix rolls out live, real-time voting features
Microsoft Copilot introduces Real Talk and video generation features
Sony launches a limited-edition “Lost Starship: Marathon” game controller
Rumors You Can Just Glance At
Sony announces a joint venture with TCL
On January 20, Sony announced that it has signed a memorandum of understanding with TCL Electronics to establish a joint venture focused on home entertainment. Under the agreement, TCL will hold a 51% stake in the new company, while Sony will hold 49%. The joint venture will fully take over Sony’s home entertainment business and will be responsible worldwide for the development, design, manufacturing, sales, logistics, and customer service of televisions and home audio products. The products will continue to use the Sony and BRAVIA brands.
Sony and TCL plan to sign the final agreement by the end of March 2026, with the new company expected to begin official operations in April 2027. Source
Ayaneo releases the Konkr Fit handheld
On January 20, Ayaneo’s sub-brand Konkr unveiled its first Windows-based handheld, the Konkr Fit. The device is powered by the AMD Ryzen AI 9 HX 470 processor, featuring the Zen 5 architecture and RDNA 3.5 graphics cores. The display has been upgraded from the previous Android model’s 6-inch panel to a 7-inch OLED screen. It packs an 80Wh battery, larger than those in the Legion Go 2 (74Wh) and Legion Go S (55.5Wh). The control layout includes Hall-effect joysticks, adjustable triggers, and dual rear buttons. On the top are two USB-C ports, while the back features a large cooling air intake and exposed screws, and it will be available in Retro Gray and Yellow color options.
Pricing and release date have not yet been announced. Source
RedMagic launches the RedMagic 11 Air and multiple new products
On January 20, RedMagic officially unveiled the RedMagic 11 Air along with several new products for its gaming ecosystem. The RedMagic 11 Air is powered by the Qualcomm Snapdragon 8 “Ultimate” Edition processor and RedMagic’s in-house Red Core R4 gaming chip. It features LPDDR5X ULTRA memory and UFS 4.1 storage, and comes with a 6.85-inch 1.5K under-display camera full-screen display supporting a 144Hz refresh rate and a 2500Hz instant touch sampling rate. The display also supports 2592Hz high-frequency PWM dimming and DC dimming. The phone packs a 7000mAh battery with 120W fast charging. Its cooling system consists of an active fan spinning at up to 24,000 rpm and a 0.5mm VC vapor chamber panel. The body also integrates 520Hz gaming shoulder triggers and an X-axis linear motor.
RedMagic 11 Air
Several other gaming ecosystem products were also announced at the event, including the RedMagic 11 Pro+, which features a carbon-fiber body, a sapphire glass back, and a “Golden Flow” liquid cooling system, as well as the RedMagic Gaming Laptop 16 Pro Golden Saga · 3D Explorer Edition, equipped with an NVIDIA RTX 5090 GPU and supporting glasses-free 3D display technology. The RedMagic 11 Air starts at CNY 3,699, while the RedMagic 11 Pro+ Golden Saga 24GB + 1TB version is priced at CNY 9,899.
RedMagic 11 Pro+
In addition, the RedMagic Gaming Tablet 3 Pro Golden Saga is expected to go on sale by the end of January, with pricing yet to be announced. Source
Zhipu releases and open-sources the GLM-4.7-Flash model
On January 20, Zhipu officially released and open-sourced the GLM-4.7-Flash hybrid reasoning model. The model has a total of 30 billion parameters, with 3 billion active parameters, and is positioned as a lightweight deployment option that balances performance and efficiency. It is now available on Zhipu’s open platform BigModel.cn with free access. In mainstream benchmarks such as SWE-bench Verified and τ²-Bench, the model’s overall performance surpasses gpt-oss-20b and Qwen3-30B-A3B-Thinking-2507, achieving open-source SOTA scores among models of the same size class.
GLM-4.7-Flash is primarily optimized for programming scenarios, covering both front-end and back-end development tasks, and is also suitable for general-purpose use cases such as Chinese writing, translation, long-text processing, and emotional role-playing. With the release of the new version, the previous free model GLM-4.5-Flash is scheduled to be officially retired on January 30, 2026, at which time all related API requests will be automatically routed to GLM-4.7-Flash. This iteration aims to provide developers with higher-density intelligent service support by improving logical reasoning capabilities and optimizing parameter scale. Source
Netflix rolls out live, real-time voting features
On January 20, Netflix officially launched real-time interactive voting for live content. The feature дебuts with the talent show Star Search, allowing subscribers to participate in multiple-choice voting or star-based ratings via their TV remote controls or the mobile app. It supports real-time global vote aggregation with time-limited cutoffs, enabling viewers to directly influence the narrative progression of live broadcasts.
Netflix said that the interactive framework was previously technically validated in August 2025 with The David Chang Live Dinner Show, and that a full rollout was confirmed at TechCrunch Disrupt 2025. Source
Microsoft Copilot introduces Real Talk and video generation features
On January 20, Microsoft Copilot rolled out the Real Talk interaction mode, which aims to deliver a more human-like and interactive conversational experience. It introduces multiple levels of depth for reasoning and writing style options, supports viewing chains of thought and reasoning paths, and offers long-context memory capabilities. During conversations, it can reference past interactions and user background, and will at times show curiosity or proactively challenge logical inconsistencies.
In addition, Copilot is testing a video generation feature on Android, allowing users to generate video clips of up to 8 seconds with audio. The underlying model for this feature has not yet been disclosed and is currently being rolled out gradually, with no mandatory subscription requirement at this stage. Source
Sony launches a limited-edition “Lost Starship: Marathon” game controller
On January 19, Sony announced that alongside the March 5 release of Bungie’s Lost Starship: Marathon, it will roll out two limited-edition hardware products: a DualSense Limited Edition controller priced at USD 84.99 and a Pulse Elite Limited Edition headset priced at USD 169.99.
The DualSense Limited Edition controller is deeply inspired by the world of Lost Starship: Marathon. Bungie’s design team said their goal was to create a “real-world object” that feels as if it were taken directly from the game itself. To reflect the title’s distinctive industrial aesthetic and environmental architecture, the controller features bold graphic elements and striking color combinations. The Lost Starship: Marathon Limited Edition Pulse Elite headset follows the same design language as the controller. Source
Rumors You Can Just Glance At
According to a January 19 post by LeicaRumors, Leica is planning to release its first 35mm focal-length lens in the Noctilux series, the Noctilux-M 1.2/35 ASPH., on January 29, 2026. This would mark the first time the series enters the 35mm documentary-style focal length category. Leaked spy photos show that the Noctilux-M 1.2/35 ASPH. continues Leica’s classic industrial design language, featuring a relatively compact aluminum body accented with the brand’s signature yellow scale markings. Notably, the lens is equipped with a fixed lens hood that users must extend by rotating it into position. Source
On January 20, Samsung published and then withdrew an upgrade announcement for Bixby on its official website. According to the notice, the new version of Bixby will deeply integrate Perplexity AI to enable web-based, real-time natural language Q&A and information retrieval, while also introducing a new device agent architecture capable of recognizing non-specific user intents and automatically triggering the corresponding system settings. The updated assistant is set to enter testing as part of the One UI 8.5 Beta program and is planned to debut as a core preinstalled feature on the Galaxy S26 series. Source
National rollout of the medical insurance drug price comparison mini program
New regulations ban the sale of 13 categories of food in livestreams
Setapp abandons plans to open a third-party iOS app store in Europe
NVIDIA blog typo misstates a unit, correction triggers copper price fluctuations
Huawei smartphone shipments return to No.1 in China for the first time in five years in 2025
Rumors You Can Just Glance At
ChatGPT to introduce ads
On January 16, OpenAI announced plans to begin testing ads in ChatGPT over the coming weeks. The initial rollout will target users in the United States, before gradually expanding to other regions worldwide. Ads will primarily appear in the free version and the low-cost “Go” plan priced at USD 8 per month (which has been tested in India for several months and will be rolled out to more markets). Users on Plus, Pro, and Enterprise tiers will not see ads for now.
OpenAI stated that ads will not interfere with or alter the content of ChatGPT’s generated answers. They will appear as clearly labeled standalone boxes placed below the chatbot’s replies. For example, when a user asks for travel advice for New York, ChatGPT will first provide a standard itinerary, and may then display ads for local hotels underneath. The head of OpenAI’s app division also said the company will explore more interactive ad formats in the future, such as allowing users to ask questions directly about the advertised content to assist with purchase decisions.
Regarding privacy concerns, OpenAI emphasized that it will not sell user data to advertisers, nor will it disclose specific conversation content. Advertisers will only receive aggregated performance metrics such as impressions and clicks. While the system will match ads based on conversation topics and some personalization data, users can disable data usage for ad targeting at any time in the settings. In addition, OpenAI has implemented strict restrictions: ads will not be shown in conversations involving sensitive topics such as health or politics, nor in chats with underage users.
OpenAI is under significant pressure to commercialize. Although ChatGPT now has more than 800 million weekly active users, the vast majority are free users who generate no direct revenue. As a company founded a decade ago with roughly USD 64 billion raised in total funding, OpenAI urgently needs to develop new revenue streams beyond subscriptions to support the high costs of model training and operations, and to meet increasingly fierce competition from rivals such as Google Gemini.
National rollout of the medical insurance drug price comparison mini program
On January 16, China’s National Healthcare Security Administration announced that the “Designated Pharmacy Medical Insurance Drug Price Comparison” mini program has been rolled out nationwide. The program integrates real-time data from designated retail pharmacies covered by medical insurance across the country, providing regularly updated information on drug prices, stock availability, and manufacturers.
According to the announcement, insured users can access the feature by searching for their local medical insurance service platform on WeChat or Alipay, or by using the official National Medical Insurance Service Platform app and locating the price comparison module. After entering a drug name, the system displays the price range at designated pharmacies in the user’s area along with detailed stock information, and supports sorting and filtering by options such as lowest price first or nearest distance. Once a pharmacy is selected, users can navigate to it directly using the built-in map function or place a call with one tap for inquiries, helping reduce unnecessary trips and avoid extra expenses.
Building on the full rollout of the basic price comparison feature, local medical insurance authorities are continuously expanding service scenarios and functionality based on user feedback. For example, to address typing difficulties among elderly users, mini programs in several provinces have added a “photo-based drug recognition” feature, allowing users to simply take a picture of a medicine box to automatically identify the drug and retrieve price comparison information. Some regions have also integrated online purchasing and home delivery services, providing added convenience for residents with limited mobility or urgent needs.
In addition, certain areas are further exploring features such as medical consumables price inquiries, prescription-based drug matching and search, and analysis of drug price fluctuation trends. These regions also display index ratings for pharmacies based on “volume–price comparisons” of insured drugs, helping the public make more informed purchasing decisions.
New regulations ban the sale of 13 categories of food in livestreams
According to Xinhua News Agency, China’s State Administration for Market Regulation recently issued the Regulations on the Supervision and Administration of Livestream E-commerce Operators in Fulfilling Their Primary Responsibility for Food Safety, which will take effect on March 20, 2026. The regulations clearly specify 13 categories of food that are prohibited from being sold via livestreaming, including: food produced using non-food raw materials or containing toxic or harmful substances; food with pathogenic microorganisms or heavy metals exceeding safety limits; expired, spoiled, deteriorated, moldy, or insect-infested food; meat and aquatic products from livestock or poultry that died of disease or poisoning, or that failed quarantine inspection, as well as their processed products; prepackaged food without labels; and food whose production or sale is explicitly prohibited by the state, among others.
The regulations bring livestream e-commerce platform operators, livestream channel operators, livestream marketers, and livestream marketer service agencies all under regulatory oversight. All of these parties are required to fulfill their respective primary responsibilities for food safety in accordance with the rules. Food producers and distributors that operate livestream channels must disclose their licensing information and verify supplier qualifications, while non-food producers and distributors are required to establish strict product selection systems. Livestream marketers must strengthen screening and vetting of products. Platforms, for their part, must set up systems for review and registration, training, and risk control, appoint food safety management personnel, formulate food safety risk control lists, and establish working mechanisms for “intelligent monitoring, investigation and scheduling, and rapid response.”
To strengthen regulatory enforcement and improve consumer protection, the regulations require market supervision authorities to include food sold via livestreaming in routine inspections and annual sampling and testing plans, and clarify that technical monitoring records may be used as electronic evidence. Platforms are also required to provide convenient channels for complaints and reports and to handle consumer requests in a timely manner.
Setapp abandons plans to open a third-party iOS app store in Europe
Ukrainian software developer MacPaw recently announced that it will officially shut down its third-party iOS app store for users in the European Union, Setapp Mobile, on February 16. On its official support page, MacPaw explained that the decision was made because the app marketplace’s “constantly evolving and complex commercial terms” no longer align with the company’s current business model, hinting at difficulties in achieving profitability.
Setapp Mobile launched a public beta in the EU in September 2024, offering iOS app distribution through a subscription model. After the service is officially discontinued, all apps obtained through the platform will be removed. The company advises users to back up important data before the deadline to avoid losing access once the service ends. The Mac version of Setapp, which is also subscription-based, will continue to operate normally and will not be affected by this adjustment to the mobile business.
Setapp Mobile came into existence thanks to the enforcement of the EU’s Digital Markets Act (DMA), which requires Apple to allow third-party app sideloading in the region. However, this emerging distribution channel faces serious challenges. In particular, Apple’s introduction of the “Core Technology Fee” rule requires apps that exceed a certain installation threshold to pay Apple a fee for each first annual installation, significantly increasing operating costs for third-party app stores and their developers.
At present, there are still five other third-party app stores operating in the EU market, including the Epic Games Store. Epic has repeatedly criticized Apple’s fee policies for hindering competitors from gaining a foothold, but it continues to operate while awaiting further scrutiny of Apple’s rules by EU regulators.
NVIDIA blog typo misstates a unit, correction triggers copper price fluctuations
According to Caixin, a blog post published by NVIDIA in May 2025 has recently drawn renewed attention from the market. In the article, NVIDIA stated that a 1-megawatt (MW) rack requires 200 kilograms of copper busbars, and that the rack busbars of a 1-gigawatt (GW) data center would require 500,000 tons (half a million tons) of copper.
Since 1 GW equals 1,000 MW, a proportional calculation would mean that a 1 GW data center should require 1,000 times 200 kilograms of copper, or 200,000 kilograms. Therefore, the “500,000 tons” figure in NVIDIA’s original text was clearly a typo. NVIDIA later corrected the mistake.
However, the figure had already been cited by many market research reports, and NVIDIA’s correction directly led to a short-term drop in international copper prices. After reaching a record high of USD 13,407 per ton on January 14, LME copper futures retreated for two consecutive days, with a cumulative pullback of about 3.4%, temporarily falling below the 10-day moving average (MA10).
In fact, driven by factors such as tariff arbitrage in the United States, constrained copper mine supply, and new demand from AI, copper prices had repeatedly hit new highs in the second half of 2025. LME copper rose from around USD 9,900 per ton in early September 2025 and, for the first time in history, broke through USD 13,000 per ton in January 2026. Citi expects copper prices to reach USD 14,000 per ton over the next three months.
Huawei smartphone shipments return to No.1 in China for the first time in five years in 2025
According to data released by IDC and cited by Nikkei, Huawei reclaimed the top position in China’s smartphone shipments in 2025, returning to No.1 for the first time in five years. In absolute terms, Huawei shipped 46.7 million smartphones in 2025, a year-on-year decline of 1.9%. However, as vivo—the top vendor in 2024—saw a sharp drop of 6.6%, Huawei overtook it to claim first place.
Previously, Huawei had been restricted in procuring high-performance semiconductors and was unable to offer 5G, which drove consumers away. In recent years, however, the company has revived sales with its Kirin chips. The launch of its latest model, the Mate 80, in November 2025 further boosted performance, added AI features capable of automatically handling various tasks, and was priced lower than its predecessor.
Apple of the United States ranked second, with shipments rising 4% to 46.2 million units. Sales of the iPhone 17 series, launched in September 2025, were strong. Toward the end of 2025, Apple also stimulated demand by offering a 300-yuan discount on its high-end Pro and Pro Max models through official sales channels.
Overall smartphone shipments in China fell 0.6% in 2025 to 284.6 million units, the first year-on-year decline in two years. Although government subsidies encouraging trade-ins provided some support, in certain regions the subsidy quotas were used up early, weakening momentum. Against the backdrop of subdued consumption, IDC forecasts shipments of 278 million units in 2026, continuing to fall below the previous year.
Rumors You Can Just Glance At
On the afternoon of January 16, Jia Guolong, founder of Xibei Oat Noodles Village, announced on his personal Weibo account that he would give a comprehensive response at 10 p.m. to what he described as serious slander and defamation by Luo Yonghao against Xibei, and invited media, netizens, and relevant government departments to follow the matter. Luo Yonghao soon reposted the message, saying he would try to remain patient. However, when 10 p.m. arrived, Jia Guolong’s personal Weibo account did not publish any update. At present, the accounts of both Luo Yonghao and Jia Guolong have been muted. The response Jia mentioned was eventually published in text-and-image form on Xibei Group’s verified account (@西贝人心声). Jia stated that Luo had implied in a previously published article that Jia had colluded with relevant authorities to carry out a “cross-province arrest” against him, which Jia said maliciously incited public sentiment. He demanded that Luo explain the matter clearly and go together to the relevant government departments to verify whether there had been any report to the police or request for an arrest. The post was later deleted. Weibo CEO Wang Gaofei (@来去之间) then posted, citing the Cyberspace Administration of China’s “negative behavior list for online celebrity accounts,” saying that “in the future, if people want to engage in public disputes, they should probably do so through media interviews.” Luo Yonghao later acknowledged that he had been muted for 15 days and said he would no longer comment on the Xibei incident.
On January 15, the mainland China App Store could no longer find the solo-living safety app “死了么” (“Are You Dead Yet”), which had recently drawn attention for its unusual name, while it remained available in App Stores in other regions. Previously, on January 13, the app announced on its official Weibo account that it would officially adopt the global brand name “Demumu” in its upcoming new version. On January 14, its official account said that the previous renaming attempt had not been satisfactory and that it was soliciting creative ideas from across the internet.
This article’s outline was generated by Gemini. I completed the content creation, and the article was polished by Gemini afterward. My personal writing isn’t great, so if you’re sensitive to AI-generated content, please accept my apologies!
Introduction: “Network Speed Anxiety” and the “Bare-Bones House” of Stock Android
Friends who use Pixel phones or other near-stock Android systems are probably familiar with the joke about a “bare-bones house.” While we enjoy the purest Android experience, there are always some missing localized features that drive people crazy—and “real-time network speed display” is one of them.
In China’s complicated network environment, a speed indicator is almost a necessity for me. The awkward part is that Google seems to have never planned to support this feature at the system level. So we have no choice but to turn to third-party apps on the Google Play Store. However, after going through similar apps on the market—such as NetSpeed Indicator and Internet Speed Meter—I found that even in 2026, the experience is still far from satisfactory.
Most of these “veteran apps” have outdated UI designs. Many of their interfaces are stuck in the Android 4.4 era or the first generation of Material Design, and they always look out of place on modern Android systems. The actual speed display isn’t great either: it’s either a tiny status-bar icon with barely readable numbers and units, or a floating window plastered with the “display over other apps” permission reminder. Feature bloat is another problem—I just want a speed indicator, yet they often bundle in traffic statistics, detailed usage analysis, and a bunch of features I don’t need.
Most critically, they all share one fatal flaw: when a proxy tool is enabled, the reported network speed is wildly inflated. Pixel users have probably run into this—your real download speed is only 5 MB/s, but the floating window shows 10 MB/s.
Why Is the Network Speed in These Tools Always Inaccurate?
Simply put, this is the fault of Android’s old statistics mechanism. Traditional speed-meter apps usually read the system’s total traffic interfaces directly. When a proxy tool is enabled, data packets first pass through the physical network interface (such as wlan0) and are counted once. Then the data is unpacked and forwarded to the proxy tool’s virtual interface (such as tun0), where it gets counted again.
Most legacy apps simply add up the traffic from all interfaces, which results in displayed speeds that are often double the real speed.
Fortunately, AI-assisted programming has been booming lately. Since I couldn’t find a perfect alternative, an idea came to me: why not let AI help me write a network-speed app tailored specifically for Pixel?
And that’s how Pixel Meter was born.
What Makes Pixel Meter So Good?
As an app built to solve my own needs, Pixel Meter mainly addresses two core problems.
Accurate Traffic Statistics (Say Goodbye to Inflated Numbers)
Pixel Meter abandons the outdated global-statistics approach and instead uses the new API TrafficStats.getRxBytes(ifaceName) introduced in Android S (API 31).
This API allows an app to precisely obtain traffic data from a specific network interface. With a built-in whitelist and blacklist mechanism for network interfaces, Pixel Meter can intelligently filter out duplicate data from virtual interfaces created by proxy tools. Most importantly, this is achieved without requiring root access or Shizuku permissions—making it both efficient and accurate.
Possibly the Most Elegant Display Method Available
The Live Update Notification is, in my view, Pixel Meter’s biggest secret weapon. On Android 16 and some other high-version systems that support this feature, live update notifications avoid the space constraints and poor readability of traditional status-bar icons, allowing much richer information to be displayed more flexibly.
Let’s look at a comparison:
Traditional approach: a crowded status bar where the numbers and units are hard to read
Pixel Meter, on the other hand, takes advantage of the notification area’s real-time update capability. Although there is a character limit (7 characters), it’s more than enough to clearly show “number + unit”:
Live update approach vs. floating window
More importantly, it doesn’t look like a third-party patch at all. Instead, it feels like a native system feature—clean, natural, and seamlessly integrated.
Behind the Scenes: I Provided the Ideas, AI Wrote the Code
The arrival of the AI era has completely changed the barrier to entry for individual developers.
In the past, I tried building a few small tools myself (such as an app to automatically skip ads), but I often gave up halfway because of the tedious initial setup and the long wait for positive feedback. The effort required was too high, the results too slow, and enthusiasm was easily drained.
Another incubation project that went from full of excitement to something I never wanted to open again
But this time was completely different. During the development of Pixel Meter, I adopted a new “human–AI collaboration” model:
I handled: architecture design, requirement analysis, and “code reviews” plus course correction when the AI made mistakes.
Gemini handled: writing the concrete implementation code, generating documentation, and even preparing store-listing materials.
I used AntiGravity together with a custom GEMINI.md rules file. It’s like giving the AI a form of “long-term memory,” allowing it to remember our previous architecture discussions and coding standards. Even in a new session, it can read these “memory files” and quickly get back into development mode.
Letting AI recommend and decide on the tech stackArchitecture design discussionsReading the “memory” files when starting a new session allows Gemini to quickly “recall” the project’s requirements.
The final numbers were surprising: Pixel Meter has already released three versions, and about 90% of the code was generated by Gemini. Even the GitHub repository’s README, the privacy policy, and the promotional copy and featured images on Google Play were all created by it.
You’re Invited to Try It Out
At the moment, Pixel Meter is open-sourced on GitHub. You can also head to the Google Play Store and search for “Pixel Meter” to download and use it directly.
Whether you’re looking for a clean, no-frills network speed indicator, or you’re simply curious about the code quality of AI-assisted development, you’re more than welcome to give Pixel Meter a try.
DJI releases the DJI RS 5 lightweight commercial gimbal
MediaTek announces the Dimensity 9500s and Dimensity 8500 chips
The Philippines grants visa-free entry to Chinese travelers
Qwen announces the launch of AI-powered lifestyle shopping features
Apple announces Apple Pay support for Visa cards
Google Gemini now rolls out “Personal Intelligence”
Microsoft to remove the Collections feature from Microsoft Edge
iQOO releases the iQOO Z11 Turbo smartphone
On January 15, iQOO officially unveiled the iQOO Z11 Turbo smartphone, with a starting price of RMB 2,699, or RMB 2,039.15 after national subsidies.
In terms of display, the iQOO Z11 Turbo features a 6.59-inch flat OLED screen with a resolution of 2750 × 1260 and a maximum refresh rate of 144 Hz. The panel uses TCL CSOT C9+ luminous material, reaching a local peak brightness of up to 5,000 nits and a minimum brightness of around 1 nit. For dimming, it supports DC-like dimming as well as high-frequency PWM dimming up to 4,320 Hz, with DC-like dimming available across the full brightness range. The display also offers up to 3,200 Hz instantaneous touch sampling and 300 Hz ten-finger touch sampling, supports 1.07 billion colors, and is protected by Schott Diamond Shield glass.
In terms of performance, the iQOO Z11 Turbo is powered by the Qualcomm Snapdragon 8 Gen 5 processor and is paired with a self-developed auxiliary chip, Q2, dedicated to gaming performance scheduling and display optimization. The device uses LPDDR5X memory and UFS 4.1 storage, along with a large-area VC liquid-cooling system. Official benchmark scores exceed 3.59 million points. In gaming tests, mainstream open-world mobile games achieve an average frame rate of around 60 fps, with overall power consumption controlled at approximately 4.54 W.
For battery life, the new model is equipped with a 7,600 mAh battery using a second-generation semi-solid-state design. According to iQOO, the battery maintains relatively stable discharge performance in both high- and low-temperature environments. Charging-wise, the iQOO Z11 Turbo supports 100 W wired fast charging and offers a direct power supply mode for gaming while charging. The cooling system uses a multi-layer heat-dissipation structure to lower core temperatures and improve heat distribution under heavy loads.
In imaging, the iQOO Z11 Turbo is the first in the Z series to feature a 200-megapixel main camera, supporting 4× lossless zoom and covering commonly used portrait focal lengths such as 50 mm and 85 mm, as well as multi-focal-length Live Photo shooting. The front camera is 32 megapixels and supports 0.8× wide-angle capture.
In terms of appearance and design, the iQOO Z11 Turbo is available in four colors: Canglang Floating Light, Halo Pink, Sky White, and Polar Night Black. The Polar Night Black version uses a fiberglass back, while the other colors use glass backs. The phone adopts an aluminum alloy frame, measures about 74.42 mm in width and 7.9 mm in thickness, weighs around 202 g, and supports IP68/IP69 dust and water resistance.
For connectivity and software, the device integrates a multi-antenna design to improve connection stability in complex network environments. On the software side, the iQOO Z11 Turbo comes preinstalled with OriginOS 6, introducing new smoothness optimizations and animation effects, along with multiple AI features for everyday scenarios such as search and sharing. Source
DJI releases the DJI RS 5 lightweight commercial gimbal
DJI today officially unveiled its new lightweight commercial gimbal, the DJI RS 5, priced at RMB 3,099 for the standard version and RMB 3,899 for the combo kit.
According to DJI, the RS 5 introduces a brand-new RS Enhanced Intelligent Tracking Module, expanding tracking subjects from people to multiple categories such as vehicles and pets. DJI claims the maximum recognition distance for human tracking reaches up to 10 meters, and the system can re-lock onto the subject even if it briefly leaves the frame. The module attaches magnetically and supports tap-to-select or box-select tracking on the touchscreen. Combined with assisted composition, it significantly lowers the barrier for complex camera movements.
In terms of stabilization and power, the DJI RS 5 delivers up to a 50% increase in peak motor torque compared to the previous generation. Together with the fifth-generation RS stabilization algorithm, it provides more stable footage during fast rotations, motion shooting, and vertical filming scenarios. On the control side, the new model supports a native electronic control handle adapter, making one-handed and low-angle shooting easier, and adds a Z-axis stabilization indicator to provide real-time feedback on vertical shake to help users adjust their walking pace.
Battery life and body design have also been enhanced. Charging speed is increased by 60%, allowing a full charge in about one hour. The standard battery offers around 14 hours of runtime, which can be extended to up to 30 hours with the BG70 high-capacity battery grip. The gimbal weighs approximately 1.46 kg, supports third-generation native horizontal/vertical switching, and has a maximum payload of 3 kg, covering most mainstream mirrorless camera and lens combinations.
In terms of expandability, the DJI RS 5 natively supports the Focus Pro motor and the DJI SDR video transmission system, features a built-in RSA communication interface, and is compatible with a wide range of official and third-party accessories. DJI also opens up the DJI RS SDK, allowing developers to customize additional professional features. Source
MediaTek announces the Dimensity 9500s and Dimensity 8500 chips
On January 15, MediaTek announced the Dimensity 9500s and Dimensity 8500 chipsets. Both chips are deeply optimized at the hardware level for generative inference and multimodal models, with native support for major global large language models (LLMs / MLLMs) and Stable Diffusion image generation models. They also introduce AI-powered ultra-clear telephoto algorithms, the Dimensity AI semantic segmentation engine, and AI reflection and flare removal technology. In addition, the chips support on-device AI enhancements for Live Photos and photo editing, as well as AI-powered summaries for calls, meetings, and document content based on local AI computing capabilities.
The Dimensity 9500s is built on TSMC’s second-generation 3nm process, integrating more than 29 billion transistors. It features a flagship all–big-core CPU architecture and includes a Cortex-X925 ultra-big core. MediaTek states that, combined with the second-generation Dimensity scheduling engine and advanced memory compression technology, the chip delivers significant improvements in performance scheduling efficiency and app launch speeds. The Dimensity 9500s also supports ray tracing, 8K HDR video, on-device AI computing, as well as 5G and Wi-Fi 7 connectivity.
Targeting the “light flagship” segment, the Dimensity 8500 also adopts a second-generation all–big-core CPU architecture and is manufactured using TSMC’s N4P process. CPU performance is up 7% over the previous generation, GPU performance is improved by 25%, and it comes with a quad-channel memory configuration. The Dimensity 8500 likewise supports ray tracing and further enhances AI capabilities for voice and imaging. Source
The Philippines grants visa-free entry to Chinese travelers
The Philippine Department of Foreign Affairs has announced a visa-free policy for Chinese citizens. Starting from January 16, 2026, Chinese passport holders will be allowed to enter the Philippines without a visa for stays of up to 14 days. This policy applies only to tourists entering through Manila and Cebu airports, and the 14-day stay period is non-extendable. Source
Qwen announces the launch of AI-powered lifestyle shopping features
On January 15, the Qwen app announced full integration with Alibaba ecosystem services including Taobao, Alipay, and Taobao Instant Delivery, opening beta access to AI-powered shopping and lifestyle services for all users.
According to the official introduction, the Qwen app now enables unified operations such as ordering food delivery, AI-assisted shopping, and booking flights directly within the chat interface. More than 400 new features have been launched, with deep integration into Alipay’s government services and Fliggy’s travel services, along with the release of a complete feature list. A new “Task Assistant” feature has also been added to support intelligent planning and execution of multi-step complex tasks. Source
Apple announces Apple Pay support for Visa cards
On January 15, Apple announced the expansion of Apple Pay’s cross-border payment support. When traveling abroad, users in mainland China can use locally issued Visa credit and debit cards to make payments at offline merchants and online scenarios that support contactless payments.
At present, Visa credit cards issued by Industrial and Commercial Bank of China, Bank of China, Agricultural Bank of China, Bank of Communications, China Merchants Bank, China CITIC Bank, Ping An Bank, and Industrial Bank, as well as Visa debit cards issued by China CITIC Bank, already support this feature. After adding these cards to the Apple Wallet app, users can make cross-border payments via Apple Pay.
In addition, Visa credit cards issued by institutions such as Shanghai Pudong Development Bank, China Construction Bank, China Minsheng Bank, and China Everbright Bank are expected to join the list of supported cards in the coming months. Mastercard has also stated that it plans to enable Apple Pay support for Chinese cardholders of some issuing banks in the coming months. Source
Google Gemini now rolls out “Personal Intelligence”
On January 14, Google announced that a new feature called “Personal Intelligence” has been opened for testing to individual accounts. This feature can integrate information from apps such as Gmail and Google Photos, helping Gemini understand contextual relationships without explicit instructions, enabling the chatbot to comprehend user data across apps and provide responses that better match personal situations.
The feature will be rolled out first to Google AI Pro and AI Ultra subscribers in the United States, and will later be added to Google Search’s AI Mode. To reduce potential risks, “Personal Intelligence” is turned off by default.
Josh Woodward, Vice President of Google Labs and the Gemini app, said that the test version may still produce judgment errors and encouraged users to actively provide feedback. In scenarios involving changes in relationships or complex personal interests, Gemini may still struggle to accurately grasp timing and context. In sensitive areas such as health, Gemini will not make proactive inferences and will only discuss based on existing data when users explicitly ask questions.
In addition, Google stated that it will not directly use users’ Gmail content or photo libraries to train models. Only some interaction data, such as user prompts and model responses, will be used to gradually optimize feature performance. Source
Microsoft to remove the Collections feature from Microsoft Edge
Microsoft recently notified users in the latest Microsoft Edge Dev build that it plans to remove the “Collections” feature from the browser. After the change is implemented, users will no longer be able to add new content to Collections.
For existing data, Microsoft is offering limited migration options. Users can move all web pages saved in Collections to Favorites (bookmarks), but this method will only preserve the page links and will not migrate any images or notes previously added. To fully retain images and notes, users must manually export their Collections data as a CSV file for offline storage. Microsoft warns that if the export is not completed in advance, the related data will be removed from user accounts later and may be permanently lost.
According to public information, the Edge Collections feature was first introduced in 2020, allowing users to save web pages, images, and notes in one place, commonly used for travel planning, research organization, and shopping lists. At present, Microsoft has not issued an official announcement regarding the removal of this feature. However, as the notice has already appeared in the Dev version, users are advised to back up their data as soon as possible to prepare for potential future product changes. Source
Tesla launches the 2026 7-seat Model Y in the U.S.
Adobe Firefly rolls out the GPT-Image 1.5 model
Nintendo denies using AI in its advertisements
Rumors You Can Just Glance At
Apple launches Apple Creator Studio
On January 13, Apple announced that it will roll out a creative software subscription bundle called Apple Creator Studio on January 28. The bundle brings together Final Cut Pro, Logic Pro, and Pixelmator Pro for Mac and iPad; Motion, Compressor, and MainStage for Mac; as well as advanced features across the iWork suite (Keynote, Pages, Numbers, and Freeform). It is designed to provide professional-grade software support and AI assistance for video editing, music production, graphic design, and productivity workflows. After the subscription launches, Mac users will still be able to purchase standalone apps such as Final Cut Pro and Logic Pro outright from the App Store with a one-time payment.
Apple Creator Studio is priced at RMB 38 per month or RMB 380 per year, with a one-month free trial. An education discount is available for college students, priced at RMB 18 per month or RMB 180 per year. Source
Insta360 releases the Link 2 Pro series webcams
On January 13, Insta360 announced the Link 2 Pro series webcams, including the Link 2 Pro and Link 2C Pro. Both models feature a 1/1.3-inch large sensor, support 4K ultra-high-definition resolution and dual native ISO technology, and are equipped with directional microphones for clear audio capture in open offices or noisy outdoor environments. They also support the Link Controller, enabling one-click activation of an AI-powered natural depth-of-field effect. The Link 2 Pro further adds AI tracking, a two-axis gimbal, and automatic framing.
The Insta360 Link 2 Pro series can integrate with Wave to deliver an all-in-one audio and video solution, automatically identifying and switching between speakers in multi-person meeting scenarios. In terms of pricing, the standard Insta360 Link 2 Pro is priced at RMB 1,758, while the standard Insta360 Link 2C Pro is priced at RMB 1,398. Source
Tesla launches the 2026 7-seat Model Y in the U.S.
On January 13, Tesla introduced the 2026 Model Y to the U.S. market. The high-end Premium all-wheel-drive version can be configured with a 7-seat layout, while the other variants remain 5-seaters.
The interior space of the 2026 Model Y is unchanged from the previous version, but a child seat has been added in the trunk area as part of the 7-seat option, priced at USD 2,500. The vehicle starts at USD 51,490. In addition, the Premium trim now includes the previously top-tier features of a high-resolution 16-inch infotainment display and a black headliner as standard, at no extra cost. Buyers of the Premium version can also opt for 20-inch graphite-colored Helix wheels for an additional USD 2,000. Source
Adobe Firefly rolls out the GPT-Image 1.5 model
On January 13, Adobe announced that OpenAI’s GPT-Image 1.5 model has been added to its Firefly AI creation platform. Firefly Pro and Premium subscribers can use this model to generate images with no usage limits until January 15. At present, the Adobe Firefly platform has integrated image, audio, and video models from Adobe itself as well as from major providers including OpenAI, Runway, Black Forest Labs, Pika, Ideogram, and Google.
The platform operates on a “generative credits” subscription system. The Standard plan costs USD 9.99 per month and includes 2,000 generation credits; the Pro plan costs USD 19.99 per month and provides 4,000 credits; and the Premium plan costs USD 199.99 per month, offering up to 50,000 credits. Source
Nintendo denies using AI in its advertisements
Following earlier claims that AI generation tools were used in its new My Mario children’s toy advertisement, Nintendo has recently denied the allegations, stating that no AI was involved in the production or filming of the ad.
Screenshot from the controversial advertisement
The controversy mainly centered on the model’s thumb shown in the image. However, the model involved told IGN in an interview that they did participate in the audition and selection process and were actually filmed for the commercial. Some netizens later added that the part in question in the image was also a normal appearance of a double-jointed thumb. Source
Rumors You Can Just Glance At
According to a report by Reuters citing Nikkei Asia, Google plans to move the R&D and manufacturing of all Pixel models except the A series out of China. Source
A recent market research report released by Counterpoint shows that Apple has taken the top spot with a 20% share of the global smartphone market, followed by Samsung and Xiaomi. Source
According to an exclusive report by Tech Planet, Pinduoduo is currently testing a new business section called “100-Billion Supermarket,” which covers multiple categories including fruits and vegetables, nuts and snacks, and dairy and beverage products. Notably, users can start claiming various instant-discount or coupon vouchers every Monday at midnight to purchase discounted items. Source
Leaks also suggest that Honor will soon collaborate with the POP MART IP and plans to launch a POP MART co-branded Honor 500 smartphone next week. Source
Google unveils the Universal Commerce Protocol (UCP)
Apple and Google reach a partnership on AI models
Xianyu launches a collectible consignment service “Yuligou”
Anthropic announces the launch of an AI healthcare service
OpenAI acquires AI healthcare company Torch
Google removes AI-generated summaries for searches involving specific medical keywords
Instagram mistakenly sends millions of password reset emails due to a system bug
China’s first 3D-printed aircraft engine completes its maiden flight
Rumors You Can Just Glance At
Google unveils the Universal Commerce Protocol (UCP)
Recently, Google, together with retail giants such as Walmart, Shopify, and Etsy, introduced a new standard for AI shopping agents called UCP (Universal Commerce Protocol), which was officially announced at the NRF (National Retail Federation) conference in the United States. UCP is a new open standard designed to unify digital commerce. Google plans to integrate UCP-related services into its search engine and Gemini in the future, enabling users to find desired products more accurately and quickly when searching or interacting with AI, and to complete purchases directly via Google Pay without leaving the interface. It is also worth noting that at the same conference, Walmart officially announced a partnership with Google that will allow consumers to shop for products at Walmart and Sam’s Club directly through Gemini. This service will launch first in the United States and will later expand to global markets. Source
Apple and Google reach a partnership on AI models
On January 12, Google and Apple released a joint statement announcing that the two companies have entered into a long-term collaboration. The next generation of Apple Foundation Models will be built on Google’s Gemini large language model and cloud technologies. These models will power Apple Intelligence features as well as a more personalized version of Siri scheduled to launch later this year. The statement said that after careful evaluation, Apple believes Google’s AI technology is the best fit for Apple Foundation Models, and that Apple Intelligence will not alter its existing privacy commitments, including on-device processing and Private Cloud Compute. Previous reports claimed that Apple pays Google about USD 1 billion annually to integrate Gemini, and that the new Siri model would have around 1.2 trillion parameters. After the announcement, Elon Musk commented on X, saying that given Google also owns Chrome and Android, this collaboration looks like an “unreasonable concentration of power.” Source
Xianyu launches a collectible consignment service “Yuligou”
Recently, Xianyu has “quietly” rolled out its new trendy-toy consignment service, Yuligou. Under this program, Xianyu officially inspects the seller’s items. Yuligou strictly controls goods shipped directly from official sources to ensure authenticity and traceability. Items are either ordered by sellers through official brand channels (such as official apps, mini programs, or Tmall flagship stores) and shipped to the warehouse by changing the delivery address, or shipped directly into the warehouse by Xianyu-certified brand partners. From receiving and quality inspection, warehousing and listing, to order picking, packing, and shipping, the entire in-warehouse process is recorded on video, with professional procedures providing peace of mind. To avoid players relying on “luck-based blind draws,” all items sold on Yuligou are pre-inspected confirmed units rather than blind boxes. In addition, sellers set their own prices. The platform compares listings and displays the current lowest price for each item in real time, making it easier for buyers to compare prices and make purchases. Source
Anthropic announces the launch of an AI healthcare service
On January 11, Anthropic officially announced that its AI product, Claude, will roll out a healthcare service compliant with the U.S. Health Insurance Portability and Accountability Act (HIPAA). The service will be available to hospitals, medical institutions, and individual users for handling protected health information. Claude has also integrated multiple scientific databases and enhanced its support for biomedical research. For individual users, Anthropic will allow people to export their personal health data from apps such as Apple Health and Function Health, making it more efficient to organize medical records and share them with healthcare providers. Anthropic said that Banner Health, one of the largest nonprofit healthcare systems in the United States, already has more than 22,000 clinicians using Claude, with 85% reporting improvements in work efficiency and accuracy. The company is also collaborating with organizations including Novo Nordisk and Stanford Medicine. In addition, all healthcare-related responses are supported with citations from authoritative sources such as PubMed and the NPI Registry, and medical user data will not be used to train the model. Source
OpenAI acquires AI healthcare company Torch
On January 13, OpenAI announced its acquisition of the AI healthcare company Torch. Founded less than a year ago, Torch has released an iOS app designed to bring together users’ medical and health data — including hospital lab results, medication records, wearable device data, and direct-to-consumer testing services — into a single place, providing a unified “medical memory” for connected AI services. The four founders of Torch will join OpenAI. People familiar with the matter said the deal was an equity transaction valued at over USD 100 million. Source
Google removes AI-generated summaries for searches involving specific medical keywords
Recently, while testing Google’s AI summary feature, The Guardian found that for some medical-related keywords — such as several specific liver function blood test indicators — the summaries generated by Google omitted conditions that are crucial for interpreting those metrics and instead presented only numerical values. Experts interviewed by The Guardian said that some of these summaries were misleading and could cause users to misjudge their own health conditions. After the report was published, Google quietly removed the AI summary feature for the related keywords. A Google spokesperson said the company would not comment on individual removals, but that when the AI summary feature performs inappropriately, Google will make broader improvements and take action under its policy framework when necessary. Source
Instagram mistakenly sends millions of password reset emails due to a system bug
Over the past weekend, millions of users around the world received password reset emails from Instagram. Security organizations and media outlets suggested that Instagram might have suffered an information leak. Meta responded that there was a vulnerability in its system that allowed external third parties to bypass the normal verification process and submit password reset requests in bulk for some Instagram users. Because the requests appeared to be legitimate, the system automatically sent genuine reset emails to the registered email addresses of affected users. Meta emphasized that this was only a flaw that triggered the sending of reset emails and did not mean that its internal systems had been illegally accessed. The company said it has identified and fixed the issue, and users who received the emails can simply ignore them. Source
China’s first 3D-printed aircraft engine completes its maiden flight
According to People’s Daily, a turbojet aircraft engine produced using 3D printing technology and developed by the Hunan Power Machinery Research Institute of AECC (Aero Engine Corporation of China) has successfully completed a test flight in Dingbian County, Shaanxi Province. This marks China’s first domestically developed 3D-printed turbojet engine, filling a gap in the engineering application of full-engine 3D printing in the country. In the past, subtractive manufacturing required machining from a solid block of raw material, resulting in significant material waste. By contrast, 3D printing is an additive manufacturing process, with material utilization exceeding 90%. It also enables precision designs that traditional manufacturing processes cannot achieve, such as labyrinth-style cooling channels and integrated load-bearing structures, making designs more open and flexible. In addition, 3D printing technology allows manufacturers to switch between different products without adjusting production lines. Small-batch and customized production can shorten the R&D cycle by more than 30%. Source
Rumors You Can Just Glance At
At CES 2026, Qualcomm officially unveiled its Snapdragon X2 Plus chip. The chip features a third-generation Oryon CPU, an integrated Adreno X2-45 GPU, and the latest Hexagon NPU. This brings notable performance gains over the previous generation, along with a significant reduction in power consumption, and more balanced improvements in AI and ray tracing performance. According to currently leaked CineBench benchmark scores, Apple’s M4 chip still leads in single-core performance with 173 points, compared to 133 points for the X2 Plus. In multi-core performance, however, the X2 Plus narrowly edges out the Apple M4 with 1,011 points versus 993 points. Source
According to the latest support documentation released by Microsoft, Microsoft Lens has been discontinued on both Android and iOS as of January 9, 2026. Starting February 9, the app will no longer be supported and will be removed from Google Play Store and Apple’s App Store. Users will still be able to scan documents until March 9, after which new scan requests will no longer be possible, though existing scan records can still be accessed via the last signed-in Microsoft account. Microsoft recommends switching to OneDrive for document scanning. OneDrive includes a built-in scanning feature that can be accessed by tapping the “+” icon at the bottom of the app interface. However, OneDrive does not save scanned files locally, storing them in the cloud instead. Source
Google has confirmed that Chrome 150, scheduled for release later this year, will be the last version to support macOS 12 Monterey. Source
Due to Australia’s ban on social media use by minors under the age of 16, Meta has removed 544,052 underage accounts since December 11, including 330,000 Instagram accounts, 170,000 Facebook accounts, and nearly 40,000 Threads accounts. Source
According to leaker Ice Universe, Samsung will adopt a new main camera sensor, the S5KHP6, on the S27 Ultra. The sensor has the same size as the HP2 released in 2023. Source
Microsoft announced the immediate retirement of the Microsoft Deployment Toolkit (MDT), an enterprise deployment tool first released in 2003. The software will no longer be supported, its download packages will be removed, and no future compatibility updates will be provided. Source
Welcome to this episode of Pai Review. You can use the table of contents to quickly jump to the sections you’re interested in. If you’ve discovered other apps worth checking out, or topics you’d like us to cover, feel free to join the discussion in the comments.
NotePlan 3.2: Simplified Chinese support, new integrated HTML plugins, and more
Platforms: iOS / iPadOS / macOS / Web
Keywords: task boards, team collaboration
@ElijahLee: NotePlan, the all-in-one time management tool that combines tasks, notes, and calendars, has been updated to version 3.2. The update adds multilingual support, including Simplified Chinese, as well as a lock screen widget for quick capture, integrated HTML plugins, support for the Liquid Glass design style, and more.
First up is multilingual support. NotePlan 3.2 now fully supports Simplified Chinese. For an app that is both complex and highly professional, native language support makes it much easier for users to get started. The new version also adds localization for Japanese, Italian, Korean, and Russian, along with VoiceOver accessibility support.
On mobile, the update introduces a new lock screen widget focused on quick capture. You can add tasks, start voice transcription, or create a note directly from the lock screen.
This lock screen widget supports custom configuration. In the widget settings, you can choose where the note content is saved and under which heading. In terms of content, the editor supports rich text formatting such as tasks, bullet points, quotes, and more. Functionally, it supports multiple input methods including voice transcription, typing, handwriting, and drawing.
On macOS, NotePlan has introduced integrated HTML plugins and provides multiple plugin templates and tutorials to help users get started quickly. In “Settings – AI & Plugins,” you can manage all plugins, download updates, and use them.
The main integrated HTML plugins in this update are the Dashboard plugin and the Linear Calendar plugin. The Dashboard plugin is a dashboard-style plugin that displays statistics, summary views, task progress, and more in the sidebar, allowing you to quickly review your work status. The Linear Calendar (calendar view plugin) provides an enhanced calendar view that integrates both the macOS calendar and the NotePlan calendar.
HTML plugins run directly within the main NotePlan window. When you open an HTML plugin, it automatically appears in the sidebar, and you can switch to it at any time just like switching between notes, calendars, or tasks. This makes HTML-based plugins much more practical in daily workflows, without the need to open a separate window.
NotePlan officially mentioned that more plugins are in development, but users can also follow the tutorials to quickly create their own plugins using Claude Code or Cursor—even without prior coding experience.
In NotePlan 3.19, the app added support for the Liquid Glass design style. The sidebar becomes floating and shows the background color of notes through it, the command bar and menus are updated with a more transparent glass-like texture, and the overall icon set is more unified and modern.
You can download NotePlan for free from the App Store. With a paid subscription, you can unlock features such as cross-platform sync, Markdown export, and integrated HTML plugins. Pricing is USD 12 per month or USD 99.99 per year. Setapp users can get it for free.
Typeless: Making Voice Input Truly Practical
Platforms: iOS / macOS / Windows
Keywords: voice input, AI transcription
@Vanilla: Most voice input methods on the market (such as WeChat Input, Doubao Input, iFlytek Input, etc.) focus on reproducing human speech with 100% accuracy, but overlook the fact that people don’t speak perfectly—they make slips, pause, and stumble. Typeless changes this. Instead of merely aiming for “accurate transcription of sound,” it aims to “reconstruct the original expression in the mind.”
Before getting into its advanced features, I’d like to first introduce Typeless’s solid basic capabilities, because these are the foundation that keeps me using it.
First is mixed-language input. Typeless supports more than 100 languages, and its recognition of mixed Chinese and English—what I use most often—is excellent. There’s no need to switch keyboards or languages; it just works.
Next are the practical built-in keys. The Typeless input interface includes @, space, delete, and a Return key. The Return key is context-aware: in WeChat it means “send,” in documents it’s “new line,” and in search fields it becomes “search.”
Then there’s the personalized dictionary. Typeless lets you create custom entries manually or by voice, accurately matching rare words or proper nouns. This ensures accuracy in the output and saves you from tedious corrections.
Finally, there’s whisper recognition. Typeless allows input by speaking very softly. For introverts like me, it can accurately recognize quiet whispers into the microphone in public places, avoiding “social death.” Typeless also works while media is playing—for example, I can dictate while listening to music with AirPods. It doesn’t interrupt playback; the music just becomes background audio and doesn’t affect recognition accuracy.
If the basic features are the skeleton, then AI is the soul of Typeless, solving the pain points of traditional voice input:
Typeless supports automatic removal of filler words. It can identify and eliminate interjections like “um,” “uh,” “that is,” as well as repeated content caused by stuttering or hesitation, keeping only the meaningful information.
Typeless supports real-time speech editing, which is its killer feature compared to other voice input methods. When I say, “Tomorrow morning at 7… ah no, let’s make it 3 p.m.,” Typeless automatically corrects this to “Tomorrow at 3 p.m.” and outputs only the final result, rather than recording the whole correction process.
Typeless supports intelligent formatting. When it recognizes lists (such as shopping lists) or multi-point content, it automatically formats them into separate lines and lists, and adapts the layout to the writing context of the app.
Typeless supports scenario-based tone polishing. It adjusts tone based on the type of app. For example, in text messages it keeps casual phrases like “kinda wanna,” while in Gmail it automatically converts them into the more formal “kind of want to.”
Typeless isn’t a flashy demo from a big tech company—it genuinely solves the problem that “voice input isn’t practical enough.” On iPhone, it’s great for replying on WeChat and other instant messaging apps, saving you from typing while still producing messages that are easy for others to read. On desktop, Typeless is well suited for writing and capturing ideas. It supports seamless switching with a physical keyboard, making it suitable as your primary input method, with the keyboard used for quick fine-tuning.
Typeless can be downloaded for free from the App Store and its official website, with a free Pro trial for the first month. Subscription pricing is USD $12 per month (billed annually) or USD $30 per month, mainly to unlock the monthly limit of 4,000 words.
TypeLouder: A Keyboard Sound Effects App Recommended by Luo Yonghao
Platform: macOS
Keywords: keyboard sound effects
@化学心情下2: Last week I recommended a mechanical keyboard and mouse click sound simulator called Funkey. This week, I’m recommending another similar keystroke sound app. This time it’s TypeLouder, which focuses more specifically on simulating keyboard sounds. The app lets you set different keyboard sound profiles and adjust volume levels, so even when typing on a membrane keyboard, you can still experience the satisfying feel of a mechanical one.
In principle, TypeLouder works much the same way as Funkey: it converts key presses into simulated mechanical keyboard sounds. The difference is that TypeLouder first asks you to choose the type of mechanical switch—such as brown, red, blue, or black switches—then set the volume. You can also preview the sound that will be produced when typing, giving you a clear idea of both loudness and tone in advance.
This approach is much more intuitive than Funkey’s method of listing a large number of keyboard model names. After all, with many modern mechanical switches, the differences in sound are often subtle, while the bigger distinctions lie in key travel stages and tactile feedback.
More importantly, TypeLouder can automatically adjust its volume based on your surroundings, so you can enjoy the intense clicking sounds without getting awkward side-eyes from people nearby.
TypeLouder also offers some thoughtful settings. For example, you can choose the audio output device (so the sounds play only in your headphones, letting you enjoy them privately without bothering others). You can decide whether the space bar or return key should produce sounds, and you can switch between different switch sound profiles using keyboard shortcuts. Overall, it’s very convenient.
@Peggy_: You leave home and realize you forgot your keys. You get to your desk and notice you didn’t bring your water bottle. You return home only to find your charger is still at the office… little things like these happen all the time. I still remember a college classmate whose behavior was legendary—almost every time he was halfway to somewhere, he’d suddenly realize he’d left something in the dorm and turn back.
PocketCheck is designed to solve exactly these easy-to-forget little things, or to constantly remind people who are especially forgetful. Through scheduled prompts and persistent notifications for important items, PocketCheck provides reminders right when you need them. First, you can set up a checklist of things to carry based on your needs when going out. If your forgetfulness is really serious, you can even take photos of the items or note where they are stored. Then, you can set time-based reminders according to your usual departure times, so that before you head out, you’ll be soul-questioned: “What’s in your pockets?”
Besides serving as an item checklist, you can also get creative and use PocketCheck as a “put-things-back” list, to avoid forgetting to return items to their original places after using them. For work tasks with relatively fixed workflows, PocketCheck can also act as a supervisor for task completion. It allows you to set an “active” status for items in a list; when you open the relevant task, the app shows a persistent notification and indicates how many activities are currently active.
In terms of notifications, PocketCheck lets you enable alerts for individual lists and freely choose which days of the week to be reminded. It also supports periodic reminders, prompting you to occasionally open the app to confirm that you’ve packed your items properly or completed your tasks. Of course, you can also set quiet periods during which no reminders are received, such as while sleeping at night.
Overall, PocketCheck’s features are clearly aimed at people who are prone to forgetting things. By combining persistent notifications, interval reminders, and task activation states, it tries to help us forget less—or even not forget at all—when it comes to important matters. You can download PocketCheck from F-Droid to try it out; the app is completely free.
Danji: Schedule & Sponsored-Content Management for Creators
Platform: iOS
Keywords: media schedules, business management
@Snow: Danji is a schedule and business-collaboration management tool designed specifically for media professionals, content creators, and freelancers. With its AI engine and built-in analytics, you can quickly create schedules and sponsored-content projects from screenshots or short text, and clearly track income, payments received, and other financial details—addressing common pain points like chaotic schedules and difficult bookkeeping.
Danji’s centralized management focuses on two areas: schedules and sponsored projects. For schedule management, in addition to filling in project info, contact details, costs, publishing progress, and reimbursement status, it also distinguishes between online and offline events. For offline events, it intelligently calculates departure times based on travel methods such as flights, high-speed rail, or driving, automatically syncs with the system calendar, and provides targeted reminders. You can also place schedule widgets on your home screen for more intuitive alerts.
For sponsored-content management, the app supports full-cycle tracking from “to be executed,” to “in progress,” to “published,” making it easy to adjust plans and monitor progress. Each project can include the contract amount, payment status, commission status, and commission amount. The app then automatically calculates net income in its statistics based on payments received, giving you a clear picture of each project’s financial performance.
Danji’s most thoughtful feature is its use of AI to parse chat screenshots, itineraries, or text and automatically generate schedules or sponsored projects. Contacts, transport details, project amounts, commissions, and more can all be recognized accurately and imported quickly, saving the time required to create entries manually. The app also offers Shortcuts support, allowing you to build screenshot workflows for even higher efficiency.
At the moment, however, Danji can only recognize a single screenshot at a time. To improve accuracy, it’s recommended to import long screenshots that include full conversations with project overviews. If you only import an itinerary, Danji may mistakenly treat departure points or destinations as the project name and location—ironically reducing efficiency instead of improving it.
As a business-management tool, rich analytics are essential. Danji’s charts clearly show monthly and yearly income and expenses, as well as the proportion of online vs. offline work. They also dig deeper into payment performance by industry, brand, and agency, generating payment-rate rankings to help identify reliable partners and potential risks—enabling more refined and professional business management.
The long-standing outlining and information-organizing app OmniOutliner has been updated to a major 6.0 version, adding support for the Liquid Glass design style and introducing features such as Omni Links. Source
Instagram reportedly experienced a sensitive user data leak. The company later stated that the issue causing the system to send emails asking users to reset their passwords has been fixed, that user data was not affected, and that such emails can be safely ignored. Source
State Council launches investigation and evaluation of competition in the food delivery market
Claude Code blocks access from third-party compatible tools
Well-known open-source framework Tailwind CSS lays off a large number of employees due to the impact of AI
Large-scale internet outage hits Iran
Cloudflare fined by Italy, threatens to stop providing network security services for the Winter Olympics
iOS 26 adoption rate lower than previous versions at the same point in time
Rumors You Can Just Glance At
State Council launches investigation and evaluation of competition in the food delivery market
On January 9, the Office of the Anti-Monopoly and Anti-Unfair Competition Committee of the State Council announced that it would conduct an investigation and evaluation of market competition in the food delivery platform services sector.
A person in charge of the department said that recently, problems such as aggressive subsidies, price wars, and traffic control have become prominent in the food delivery platform services industry, squeezing the real economy and intensifying “involution-style” competition within the sector, drawing strong reactions from various parts of society. This investigation and evaluation will involve on-site verification, face-to-face interviews, and questionnaire surveys to gain an in-depth understanding of competitive practices among food delivery platforms, widely solicit opinions from platform-based merchants, new forms of workers, consumers, and other stakeholders, comprehensively assess the state of market competition, organize analysis and verification, transmit regulatory pressure, and propose measures for handling the situation.
Subsequently, Meituan and Taobao Flash Delivery both issued statements saying they would actively cooperate.
Earlier, in February 2025, JD.com entered the food delivery business, becoming the starting point of the subsidy war in the sector. At the end of April, Alibaba announced its entry into the competition and upgraded the instant retail service “Hour Delivery” under Taobao and Tmall to “Taobao Flash Delivery,” rolling out high-profile subsidies for food delivery and further announcing a RMB 50 billion subsidy plan on July 2. In response to the competition, Meituan followed suit and joined the subsidy campaign.
In this context, in May and July 2025, the State Administration for Market Regulation and other departments held two rounds of talks with Ele.me, Meituan, and JD.com. In September, the State Administration for Market Regulation organized the drafting of the Basic Requirements for the Management of Food Delivery Platform Services (Draft for Public Comment), which was implemented as a recommended national standard in December, setting out provisions on merchant management, fees and promotional practices, labor management, and dispute resolution.
Claude Code blocks access from third-party compatible tools
On January 9, Anthropic confirmed that it had deployed technical measures to prevent third-party applications—represented by popular terminal-based AI coding tools such as OpenCode—from masquerading as its official tool Claude Code to access its models, in an effort to crack down on behavior that circumvents commercial API fees.
In response, Anthropic explained that unauthorized third-party access had caused hard-to-diagnose technical failures, leading to reduced platform stability. In addition, some xAI employees were found to be using the Cursor IDE to call Claude at scale to assist in developing their own models, violating the exclusivity clause in Claude’s terms of service that prohibits using the service to build competing products.
However, community discussions generally believe that economic considerations are the primary motivation behind this ban. The Claude Max subscription plan is priced at USD 200 and offers a generous usage allowance. By comparison, achieving a similar level of compute through pay-as-you-go API calls would easily push monthly costs beyond USD 1,000.
In response to the blockade, OpenCode quickly launched a USD 200-per-month OpenCode Black plan, switching to access Claude via enterprise-grade APIs.
Well-known open-source framework Tailwind CSS lays off a large number of employees due to the impact of AI
This week, Tailwind Labs, the company behind the well-known open-source CSS framework Tailwind CSS, confirmed that it has been forced to lay off three out of four members of its development team—75% of the entire team—due to what it described as the “brutal impact” of AI. CEO Adam Wathan revealed that although usage of Tailwind CSS is growing at a record pace, the company’s revenue has plunged by nearly 80%. Without restructuring, its funds are expected to run out within six months.
Wathan pointed out that the widespread adoption of AI coding tools has changed developers’ working habits, leading to a 40% drop in visits to the official documentation site over the past two years. Tailwind Labs’ main business model is selling “lifetime license” Tailwind UI component libraries and templates to developers who visit the documentation. Now that developers can obtain code directly from AI, they no longer need to visit the website or purchase paid component libraries, making this business model difficult to sustain.
Released in 2019, Tailwind CSS is one of the most popular CSS frameworks. Many AI models are particularly good at writing Tailwind code and tend to use the framework by default.
Wathan disclosed the situation during a recent controversy in the GitHub community. Previously, some contributors proposed optimizing the documentation format to make it easier for large language models to crawl and learn from, but Wathan rejected the idea. He explained that making it easier for AI to access the content for free would make the business “even more unsustainable” and could eventually turn the project into “abandonware” due to a lack of maintainers, citing the background described above.
The layoffs sparked widespread discussion in the tech community and reignited criticism of industry giants that use open-source projects and train AI models on open-source code without giving back. Vercel and Google AI Studio later announced that they would become sponsors of Tailwind CSS.
Large-scale internet outage hits Iran
According to Xinhua News Agency, around 8:00 p.m. local time on January 8, internet services in Iran’s capital, Tehran, were disrupted. NetBlocks, an international NGO that monitors global internet connectivity, said Iran is implementing nationwide network controls related to ongoing protests in multiple regions. Cloudflare’s routing data shows that Iran’s IPv4 connections dropped sharply between the 8th and 9th before partially recovering, while IPv6 connections fell to zero. (Technically, IPv6 traffic is harder to censor than IPv4.)
The Guardian, citing expert analysis, reported that this shutdown set new records in both scale and technical measures, causing about 90% of the country’s internet traffic to vanish almost instantly. A whitelist mechanism was used during the blackout—for example, Iran’s Supreme Leader Ali Khamenei was still able to post on X.
In addition to the internet shutdown, international long-distance calls within Iran were blocked, mobile communications services were completely paralyzed, and most regions entered a digital vacuum with no signal and no service. The Starlink satellite internet system, which had helped maintain connectivity during the 2022 protests, was also subjected to targeted signal interference this time.
Cloudflare fined by Italy, threatens to stop providing network security services for the Winter Olympics
Recently, Italy’s communications regulator AGCOM fined Cloudflare €14 million (about RMB 114 million) for failing to cooperate with the country’s “Piracy Shield” system in blocking pirated content. In response, Cloudflare threatened to stop providing cybersecurity services for the upcoming 2026 Milan–Cortina Winter Olympics and is considering a complete withdrawal from the Italian market.
Under Italy’s Piracy Shield legislation, internet service providers are required to unconditionally block IP addresses and domain names suspected of infringement within 30 minutes of receiving a report from copyright holders. Since its launch in 2024, the system has been highly controversial in the tech community. Critics point out that its automated blocking mechanism lacks transparency and forces DNS resolvers to implement blocks within extremely short timeframes, making it prone to mistakenly affecting legitimate websites. For example, it previously caused large-scale internet disruptions after accidentally blocking lawful services such as Google Drive.
Cloudflare CEO Matthew Prince sharply criticized the mechanism on X, calling it a “censorship program without judicial oversight,” and accusing it of attempting to force Cloudflare’s public DNS service (1.1.1.1) to enforce blocks globally rather than only within Italy. Prince said the company will firmly appeal the decision and plans to travel to Washington and Lausanne next week to discuss the matter with the U.S. government and the International Olympic Committee (IOC), respectively.
As countermeasures, Prince listed four potential actions: terminating millions of dollars’ worth of pro bono cybersecurity support for the Winter Olympics, stopping the provision of free services to all users in Italy, removing all servers located in the country, and canceling plans to invest in setting up local offices. With the Winter Olympics set to open on February 6, a Cloudflare withdrawal at this point could expose the event to serious risks of cyberattacks.
iOS 26 adoption rate lower than previous versions at the same point in time
According to data released by web traffic analytics firm StatCounter, as of January 2026 only about 15% to 16% of active iPhones worldwide are running iOS 26, while more than 60% of devices are still on iOS 18.
Based on these figures, the adoption rate of iOS 26 appears to be less than one quarter of that of several previous versions over the same period. For example, StatCounter’s data from January 2025 showed that about 63% of iPhones were running some version of iOS 18 roughly four months after its release. In January 2024, iOS 17 reached about 54% adoption over a similar timeframe, while iOS 16 had already surpassed 60% by January 2023.
StatCounter’s estimates are derived from web traffic analysis, tracking operating system versions through page views across participating websites worldwide. Different methodologies, however, produce different results. For instance, TelemetryDeck reports iOS 26 at 60% adoption, with iOS 18 still at 37%.
Well-known Apple news site MacRumors noted that in the first week of January last year, 89.3% of its visitors were using iOS 18. During the same period this year, only 25.7% of its readers were on iOS 26. As Apple has yet to release official figures, the true penetration rate of iOS 26 remains unclear, but these data suggest that user hesitation toward iOS 26 is unprecedented in recent years.
Rumors You Can Just Glance At
According to social media posts, the chairman of China’s domestic Linux distribution UnionTech UOS reportedly fired a kernel engineer over his attire. The company recently issued a last-minute notice requiring employees to wear suits to the annual party. When a key engineer responsible for Linux kernel development asked in a group chat, “What if I don’t have a suit?”, the chairman allegedly dismissed him on the spot.
According to what some users were recently told by China Merchants Bank customer service, Visa credit cards issued by the bank can be added to Apple Pay starting January 15. Previously in mainland China, Apple Pay only supported UnionPay debit and credit cards. Apple has also updated the wording on its Apple Pay supported card list from “UnionPay credit and debit cards” to “bank cards.”
Wccftech claims that based on data from several Czech e-commerce websites, the 512 GB version of the Steam Machine is priced at about 19,826 Czech koruna (around RMB 6,622), while the 2 TB version is about 22,305 koruna (around RMB 7,450). However, retail prices for electronics in Europe usually include around 20% value-added tax, and third-party retailers often add extra channel markups.
On January 11, the team behind the recently discussed solo-living safety app “Dead Yet?” responded regarding the app’s name and the decision to start charging fees. They said they appreciate the positive suggestions from users about a new name and will carefully study and consider them. In order to keep the project healthy and sustainable, and to cover the growing costs of SMS messages, servers, and other expenses, they will introduce a paid plan priced at RMB 8. The team also welcomes interest and contact from investors, and will choose partners—institutions or individuals—who can best support the project’s growth. Finally, they called on more people to pay attention to those living alone and to offer them more care and understanding. In the “Dead Yet?” app, users must set emergency contacts and check in regularly; if they fail to check in for several consecutive days, the system will automatically send an email to the emergency contact the next day.
If last year’s CES was still half about showing off muscle and half about goofing around, then this year’s show is fully powered by the strong winds of AI. Everyone is desperately trying to “do something wild,” to the point where, without a bit of absurdity, it’s hard to leave any impression at all in the sea of people.
Some of these stunts actually come with decent ideas behind them, and a few have even turned into real products. But the overarching theme this year is obviously AI: AI can be slapped onto anything. In 2025, when everyone has outsourced their imagination to prompts, conventional products instead start to reveal a different kind of beauty—one that doesn’t rely on AI to exist.
So what flashy new gadgets did CES 2026 bring us? Here’s a roundup of some of the more unusual “memory points.”
Dell’s laptop: undoing a “betrayal of the ancestors”
Last year, Dell made what many would call a decision that betrayed its lineage: renaming its product lines to Dell, Dell Pro, and Dell Pro Max, claiming this would make it easier for customers to find the AI PC that suited them (it didn’t). Perhaps after seeing big brother Microsoft’s Copilot struggle over the past year, Dell suddenly realized that whether the name sounds like Apple’s or whether it has AI positioning doesn’t really matter—the only thing that matters is sales.
So the XPS, which had been put in the coffin for a year, was lifted back out at CES 2026.
As for the specs, nothing special—pretty standard. The design is the classic XPS look. Pricing-wise, the XPS 14 starts at USD 2,049 (about RMB 14,310), and the XPS 16 starts at USD 2,199.99 (about RMB 15,365). Dell also said an Ubuntu version of the XPS 14 will be available later, which should be a bit cheaper.
Aqara’s smart lock lets you walk in without even noticing
But you still feel tired when you get home from work, right?
At CES 2026, Aqara unveiled the world’s first Apple Home smart lock with UWB support: the Smart Lock U400. Thanks to its ultra-wideband chip, if you’re using an iPhone or Apple Watch, the door will automatically unlock as you approach—no need to say “open sesame.” Once you’re inside, it even locks itself for you.
Of course, Android users don’t need to feel left out. It also supports fingerprint, NFC, and passcode unlocking, and uses the Matter over Thread protocol, so it can join ecosystems like Google Home, Home Assistant, and Samsung SmartThings.
All in all, UWB has been in smartphones and wearables for years now, and it’s about time it unlocked more truly “frictionless” scenarios in smart homes. Hopefully, Aqara—long known for its close ties with Apple—can use this smart lock to nudge other manufacturers as well.
Sony’s console highlight wasn’t at CES
Probably feeling that the PS5 is already five years old and the market is getting a bit tired, Sony rolled out a brand-new color lineup for its controllers and consoles on January 8: The Hyperpop Collection.
In plain terms, that means limited-edition, high-saturation reskins for DualSense controllers and PS5 Slim console covers—in Nintendo red, Xbox green, and Sony blue. Patch it up, repaint it, and sell it for another year.
As for pricing, the console covers are USD 75 (about RMB 524), and the controllers are USD 85 (about RMB 594). Want a PS6? Keep dreaming. All I know is that a friend who just bought a PS5 Pro is grinding his teeth right now, because Sony almost simultaneously launched a new model called CFI-7100B01. Not many changes—just some optimizations to performance, cooling, and noise control, plus a bit more battery life for the bundled DualSense controller, while keeping the overall price the same.
He’s probably mad that Sony didn’t bother to announce something this important at CES for everyone to hear about, huh?
Lenovo squeezes an ultrawide “beltfish screen” into a laptop
At CES 2026, Lenovo showcased the Legion Pro Rollable, a gaming laptop with a roll-out display. The key feature: once you open it, you can physically adjust the screen size—from the original 16 inches horizontally expanding to 21.5 or even 24 inches.
According to Lenovo, the screen has three modes: 16 inches is “Focus Mode,” 21.5 inches is “Tactical Mode,” and 24 inches is “Arena Mode.”
Image: The Verge
But honestly, compared with hardcore gamers, I think this machine would sell better to office workers commuting on the Beijing–Shanghai high-speed rail or flights. The three sizes could just as well be renamed “Boss Mode,” “Manager Mode,” and “Corporate Drone Mode.” Also, the Legion Pro Rollable is still only a concept for now, and there’s no word yet on whether it will actually go on sale.
Intel’s new chip, supposedly with longer battery life
This year at CES, Intel unveiled its new Core Ultra 300 series processors based on the Intel 18A (2 nm) process. Intel claims that its top-end X9 388H model delivers up to 60% higher multi-threaded performance, 77% better gaming performance, and battery life extended to as much as 27 hours.
Wait a second—why did Thunderbolt 5 turn into Thunderbolt 4…
Congratulations: in the past, you could still apply for an x86 Windows laptop and use “no power on the plane” or “the computer is lagging” as excuses to delay the urgent task your boss suddenly dumped on you. In the future, you might—still be able to use the same excuses. After all, specs on paper are one thing; Windows is another.
This Dutch company is also here to grab a piece of “Teacher Luo’s underwear”
DuRoBo, a company from the Netherlands, unveiled a phone-sized (6.13-inch) e-reader called Krono at CES 2026. Judging by its looks alone, there’s nothing particularly special about it, and its specs—6 GB RAM + 128 GB storage, running Android 15—can only be described as just barely adequate.
But unlike traditional e-ink readers in the past, it integrates an AI assistant called Libby. With a press of the button on the left side of the device, you can easily record voice notes, and the AI will transcribe them into text.
You really have to admit: once you put AI into “Teacher Luo’s underwear,” it can feed quite a few tech companies.
This Samsung speaker looks like a MagSafe power bank
After getting a taste of success in home-decor-friendly products with its Frame TVs and wall-art Bluetooth speakers, Samsung this time launched a new Wi-Fi speaker lineup at CES aimed at blending into living-room interiors: Music Studio 5 and Music Studio 7.
The Music Studio 5 comes with a 4-inch woofer and a tweeter, while the Music Studio 7 features a 3.1.1-channel spatial audio system and supports 24-bit/96 kHz high-resolution audio, and can link up with other Samsung speakers for synchronized playback.
But I really want to know whether the designer of the Music Studio 7 once looked over at the desk next door during work slacking time and noticed a MagSafe power bank.
My house isn’t big, but cleaning has gone 3D
When robot vacuum cleaners were first born, they were basically round plates with a few wheels. But this design has an unsolvable problem: it can’t handle height differences or stairs. For households with split-level floors or multi-story villas, if you want to be lazy and not sweep the floor yourself, you have to buy one for every floor.
To address this, Roborock and MOVA (Dreame) offered two very different solutions—one adds legs, the other adds wings. Roborock’s G-Rover claims to be the “world’s first wheeled-legged robot vacuum,” combining multiple sensors and AI navigation to automatically climb up or down when it encounters stairs or changes in floor height.
MOVA Pilot 70 goes one step further: why not just give it wings like a drone?
While these technologies do solve a major long-standing problem for robot vacuums, in terms of their core job—cleaning—there doesn’t seem to be much new to talk about. Or is it that today’s cleaning problems are already too trivial for mopping robots to really show their skills? If you’re using one, share your thoughts in the comments.
It turns out Kindle is what puts nail technicians out of work
Anyone who gets their nails done regularly knows the drill: it takes time and damages your nails. Now iPolish claims to solve this with “digitally color-changing nails.”
Using material technology derived from Kindle e-readers, iPolish has created a nail product that can change color under certain conditions. It charges via a companion wand, pairs with your phone, and can switch among more than 400 colors in just five seconds.
Most importantly, compared with other expensive, borderline-abstract gadgets at CES, this one is actually pretty affordable. The starter kit costs only USD 95 (about RMB 663) and includes two different nail shapes, while replacement sets start at just USD 6 (about RMB 42).
This might be the product from CES 2026 that I most want to get…
AI has Tony—maybe this really works
A friend once described Google I/O by saying, “even stray cats passing by have to shout ‘AI.’” At CES 2026, the message is: even hair clippers are AI.
You’ve probably experienced getting a “bowl cut” from Teacher Tony at a barbershop. But with these clippers called Glyde, you can block Tony and cut the hairstyle you want at home. Just pick your desired style in the Glyde app, put on a positioning cap, and ask someone to hold the clippers for you. During the haircut, the AI automatically controls how far the blades extend, trimming only the parts needed according to your preset style preferences.
That said, the clippers currently seem suitable only for short hair. Glyde still can’t be used alone—you need someone else to hold them—and it’s unclear whether the app supports fine-grained settings like trimming bangs.
You don’t need a sledgehammer to kill a chicken, but I do need ultrasound to cook
This time, Seattle Ultrasonics brought us an ultrasonic knife called the C-200. When you press the orange button on the handle while cutting, the blade vibrates more than 30,000 times per second, making it much sharper—while your hand feels nothing unusual at all.
The ultrasonic knife is priced at RMB 2,712 and will start shipping in March this year.
Of course, for environmental reasons, the knife doesn’t come with a USB-C charging cable. Think wired charging for a knife isn’t elegant enough? Then you can buy a storage + wireless charging set to magnetically mount the knife on the wall, priced at RMB 3,288.
From desk setups to showing off your “partner”: AI moves onto your desktop
At CES 2026, Razer unveiled a 3D holographic AI assistant called Project AVA, which can display five different characters as holographic projections on your desk.
Razer says Project AVA is a desktop AI companion. Built on last year’s concept of an esports AI coach, it can stay by your side 24/7, “bridging the gap between virtual assistants and physical companionship.”
Similar products also exist, such as Lepro Ami from China. It doesn’t use holographic projection, but instead relies on a curved OLED display.
I can’t help but wonder: after AI takes over the world, will it start putting humans into bottles to play with too?
So… Rolex or Apple Watch? What a dilemma
According to the creator of this watch strap, David Ohayon, speaking at CES, he absolutely hated having to decide every morning in his mansion whether to wear a Rolex or an Apple Watch. Now, he no longer has to.
If you happen to own both a Rolex and an Apple Watch, this USD 400 (about RMB 2,800) strap might be worth considering. It lets you wear both watches on the same wrist—one to tell the time, the other to track your activity (and maybe check whether the more expensive watch is more accurate).
If your finances are even stronger than those of the people who buy those framed TVs mentioned earlier, you can also opt for the titanium version, priced up to USD 700 (about RMB 4,885, not including the smartwatch or the Rolex). And if you’re not using an Apple Watch, versions for Samsung, Google, and Fitbit are also available.
I think the next step for this guy is figuring out how to stop the strap from pinching arm hair on both sides.
CES’s ultimate oddball: the anti-AI lamp, Light Saver
I’ll admit that AI is indeed great and capable of many things. But when it comes to AI devices inside the home, most of them rely on cameras to collect data. Imagine if your fridge, speakers, clothes-folding robot, robot vacuum, picture frame, and desktop companion all had cameras. One day your home might get hacked—or the AI might awaken—and all these devices could suddenly “defect.”
sPresto thought about this tricky problem too, so they introduced an “anti-AI” lamp called Light Saver. It’s designed specifically to disable the cameras in these devices by emitting light that is harmless to human eyes but interferes with camera sensors, rendering them unable to capture images.
Light Saver is suitable for places that value privacy, such as homes, hotels, hospitals, and archives, helping prevent secret filming. And considering that “this year’s CES probably couldn’t even happen without AI,” sPresto’s anti-AI lamp almost feels like a sharp, rebellious outlier in an age obsessed with playing it smooth.
At this price, the frame itself is part of the art
At CES 2026, more than one company launched smart picture frame products.
For example, Fraimic introduced its Smart Canvas color e-ink smart frame, which looks almost like a physical print. The smaller size is 13.3 inches, priced at USD 399 (about RMB 2,791); the larger one is 31.5 inches, priced at USD 999 (about RMB 6,990). Add the RMB 20 print I bought on Taobao, and who’s to say what’s hanging on the wall isn’t a piece of art with some real value?
Another company, Vinabot, released a picture frame that can move—and even talk. They say the inspiration comes from the magical portraits in Harry Potter. All you need to do is upload a photo and a simple prompt, and the person in the frame can come to life and chat with you, including deceased celebrities or loved ones. As for the price, Vinabot says it will launch on Kickstarter soon, starting at USD 299 (about RMB 2,086) after a 40% discount.
Both frames can generate images directly via large AI models based on your needs.
But at this point, I don’t know why, I suddenly thought of the idle iPad Air sitting in my hands.
By the time I finish a shower, I’ve already made a fortune
There’s a common way to describe how fast Wall Street tycoons make money: in the time it takes you to do XXX, they’ve already earned an N-digit sum.
Superheat brought a hot water heater built on a similar idea to CES 2026. It doesn’t use fire, electricity, or natural gas to heat water, but Bitcoin-mining ASIC chips instead. Their thinking is that mining chips generate so much heat anyway—rather than struggling to dissipate it, why not use that heat to warm water?
What if you hit the jackpot while taking a shower?
Superheat says the installation cost is USD 2,000 (about RMB 13,967.5), and its power consumption is about the same as a regular water heater—basically free crypto mining.
To be fair, the design of this heater does look great, very Jonathan Ive-like. I’d recommend domestic manufacturers take notes.
For emotional companionship, cute may matter more than realistic
Perhaps as people have grown increasingly lonely in recent years, many products designed to provide emotional value with AI have also appeared at CES.
For instance, Takway brought an egg-shaped virtual pet, similar to a Tamagotchi. After you buy it, you first have to incubate it for two days, then it “hatches” and “grows” a cute little face. Next, you have to take care of it for a while and raise it into an adult egg—otherwise it dies and everything resets.
Once it grows up, it can take care of itself and keep you company. It will even go out on its own, then come back and tell you where it went and what fun things it did. (Hmm? Why does that sound so familiar?)
You can also buy little outfits to dress it up.
When it reaches 51 years old, it won’t die anymore—unless you stop charging it. No idea how they came up with that setting.
Another similar product is the bionic panda Anan. It comes from Shenzhen and is designed to provide companionship for the elderly and people with cognitive impairments, as well as timely reminders to prevent forgetfulness. It can also collect certain data to help caregivers understand the health condition of those being cared for.
Compared with making those very abstract humanoid robots, I think this approach is pretty good.
After all, eggs and pandas are much cuter, right?
Conclusion
That wraps up our CES 2026 roundup. Given the limited space, many major and interesting products didn’t make it in. As for some of the truly abstract, hopelessly AI-obsessed creations—like an ice maker that uses AI to reduce workplace noise—we won’t dive into those here either.
Feel free to share in the comments the CES 2026 moments that left the deepest impression on you, along with your spiciest takes. And here’s hoping next year’s CES will be even more fun.
