Alibaba

ByteDance

CATL

Huawei

Meituan

Tencent

Xiaomi

Most Followed

KrASIA Connection

Venture Roundup

Earnings Reports

China's Auto Sector

Inside China's Startups

100 AI Creators

View All

Collections

News

Features

Insights

Ecosystem

Press Releases

DeepSeek

Li Auto

Luckin Coffee

Mixue Bingcheng

Pop Mart

Xpeng

On the Radar

Events

Latest

About

Consumer

Robotics

Fintech

Collaborate

… system architectures, or handled global marketing. It’s a team that really fits the AI-plus-hardware model.
We also share a mindset. We all love creating technologies and products that users can feel and experience.
<h3>36Kr: But 2023 wasn’t exactly a great year to launch a startup. What pushed you to do it anyway?</h3>
NG: Right, 2023 was the toughest year for investment. Everyone said it wasn’t the time to start a company. But when OpenAI fired the starting gun, it signaled the beginning of a new era. Everyone, big or small, was at the same starting line. No one had prior experience. That kind of uncertainty is the best environment for new species of companies.
We eventually summarized our approach into two principles: short-term viability with long-term exponential potential, and strong technical barriers where we have an advantage.
When you apply those filters, there aren’t many options left. We explored other categories but dropped them. Either competition was too intense, or the growth logic didn’t hold.
<h3>36Kr: So NAS became the final choice?</h3>
NG: Yes. NAS is relatively easy to reason through. Data will only become more valuable as all devices eventually need to connect to a central data hub. NAS already has the computing power, bandwidth, and storage to make it the ideal platform for on-device AI.
<h3>36Kr: How did you first identify NAS as a promising direction?</h3>
NG: I talked to many peers, and several recommended NAS. I’ve been a NAS user myself for over a decade, using products from brands like ZSpace, QNAP, and Synology. Another key insight was that many DJI users are also NAS users, which helped us recruit early testers and understand their pain points.
What surprised me was how counterintuitive the feedback was. When you ask NAS users if they need AI, almost everyone says no.
<h3>36Kr: I have to admit, I was surprised too when I first heard of an AI-powered NAS.</h3>
NG: I saw the same reaction on Reddit. Most people were adamant that NAS doesn’t need AI. Some long-time users even dismissed it as marketing hype.
But think about it: if your data keeps piling up and you can’t make use of it, it’s basically wasted. AI can help you organize, manage, and even create new value from that data. From a behavioral standpoint, that logic is sound. The early negativity was just noise.
<h3>36Kr: It seems video creators would be among the first to feel the pain of storage management. Could you give an example?</h3>
NG: Sure. Video creators rarely shoot footage in sequence, so their clips are scattered and disorganized. AI can automatically recognize, categorize, and align these materials—videos, images, even scripts—saving huge amounts of time in post-production.
Another common pain point is retrieving past footage. Previously, creators had to search through hard drives manually. Now, with natural language search, they can find clips instantly. The improvement in efficiency is enormous.
<h3>36Kr: So Zettlab enables video search through AI? Traditional NAS devices can’t do that?</h3>
NG: Exactly. Let’s say you’re a sports photographer covering a game. You shoot two hours of footage and need to produce a highlight reel fast. Normally, you’d have to watch all two hours. With natural language search, it takes less than a minute to locate and clip the highlights. That’s a major leap in productivity.
<figure id="attachment_158289" aria-describedby="caption-attachment-158289" style="width: 480px" class="wp-caption aligncenter"><img loading="lazy" decoding="async" class="size-full wp-image-158289" src="https://console.kr-asia.com/wp-content/uploads/2025/11/Zettlab-Video-Production-Suite.webp" alt="A preview sequence by Zettlab, featuring its video production suite." width="480" height="270" /><figcaption id="caption-attachment-158289" class="wp-caption-text">Designed with creators in mind, Zettlab’s NAS includes a video production suite that helps streamline workflows. Graphic source: Zettlab.</figcaption></figure>
<h3>36Kr: Privacy is a major concern, especially for overseas users. How do you balance capability with data protection?</h3>
NG: That’s why we chose an edge AI architecture. All core functions run locally, without relying on the cloud. Internet connectivity is there only for convenience, say, when users want remote access while traveling.
All searching, recognition, and processing happen directly on the device. Even synchronization between devices is peer-to-peer, without passing through third-party servers. This ensures performance and privacy can coexist.
For most users, their data remains entirely on their own devices—secure, controlled, and trustworthy.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3535998221032321">article</a> was written by Leslie Zhang for 36Kr.

Can AI make NAS cool? Former DJI engineer Noah Guo built Zettlab to find out

Artificial Intelligence

Hardware

Deep Tech

China

… Bingqiang and Peak Patience Capital, with Zhuiguang Capital serving as the financial advisor. The funds will accelerate product development, team growth, and market expansion. Incubated at Xbotpark, Lissome develops robotics-driven kitchen technology that integrates smart sensing and motion control. The company holds more than 30 patents, and its latest product—a capsule-style dishwasher—combines automation with compact design for modern kitchens. —<a href="https://36kr.com/p/3543206977597316">36Kr</a></li>
</ul>
whereRelatedArticle
<h3>VoidZero, Rolo Robotics, Uluu, and more made recent headlines:</h3>
<ul>
<li aria-level="1"><a href="https://voidzero.dev/">VoidZero</a>, a Singapore-based company developing next-generation open-source JavaScript tooling, raised USD 12.5 million in a Series A round led by Accel, with participation from Peak XV Partners, Sunflower Capital, and angel investors including Koen Bok (Framer) and Eric Simons (StackBlitz).</li>
<li aria-level="1"><a href="https://www.rolorobotics.com/">Rolo Robotics</a>, a Singapore-founded robotics startup developing autonomous hot food kiosks, raised USD 3.45 million in an oversubscribed seed round backed by A2D Ventures, Beenext, Seedstars, TIS Japan, Antler, Lotus One Investment, and Blueprint Ventures.</li>
<li aria-level="1"><a href="https://www.uluu.com.au/">Uluu</a>, a Perth-based biomaterials startup, raised SGD 13.6 million (USD 10.4 million) in a Series A round led by Burda Principal Investments, with participation from Main Sequence, Novel Investments, Startmate, Fairground, and Trinity Ventures.</li>
</ul>
If there are any news or updates you'd like us to feature, get in touch with us at: <a href="mailto:hello@kr-asia.com">hello@kr-asia.com</a>.
<style>.addcaption{font-size:14px!important;line-height:20px!important;color:inherit;padding-top:8px;border-top:1px solid #e8e8e8}</style>

Deals in brief: Grab invests USD 60 million in Vay, Transcelestial secures new funding, Bioactivx closes seed round, and more

Grab

Southeast Asia

Venture Capital

Private Equity

Investment

Merger &amp; Acquisition

Singapore

Thailand

… OpenAI for its Stargate project this month, underscoring its continuing importance in the field.
Samsung's revived fortunes in the HBM segment come as the company performs strongly in another key area, smartphones, buoyed by the success of its new foldable Z series.
"We're expecting Samsung to see tailwinds on the mobile side with uplift from the strengthening mix coming from strong foldables performance, in particular, with the new Galaxy Z Fold7," Counterpoint Research vice president Neil Shah said in a note last month.
But success in HBM would be particularly sweet for Samsung as it looks to reassert its chip dominance.
Chairman Lee has attributed the company's HBM progress to two factors. One, encouraging its executives to speed up the process, and two, strengthening his personal relationship with Nvidia CEO Jensen Huang. That closeness could be seen when the two executives hugged each other at a business roundtable meeting in Washington in August. How close their companies remain could come down to Samsung continuing to keep pace in the AI memory race.
This article first appeared on <a href="https://asia.nikkei.com/business/technology/tech-asia/samsung-shakes-off-ai-memory-woes-but-race-to-innovate-is-far-from-over">Nikkei Asia</a>. It has been republished here as part of 36Kr’s ongoing partnership with <a href="https://www.kr-asia.com/nikkei-teams-up-with-36kr-in-asia-tech-news-coverage/">Nikkei</a>.

Samsung shakes off AI memory woes but race to innovate is far from over

Samsung

East Asia

South Korea

Semiconductors

Xeonova, a nuclear fusion technology company, has raised a nine-figure RMB sum in a pre-Series A funding round, according to 36Kr, citing sources familiar with the matter. The round was led by <a href="/company/ant-group">Ant Group</a>, with participation from Hidden Hill Capital, Zijin Mining, Bian Era Technology Holdings, Soul Capital, Hua Capital, Luen Mei Quantum, and Pagoda Innovation. Existing investors CMBC International and Renfa New Energy also increased their stakes.
The funds will be used to enhance the performance of Xeonova’s ongoing reactor project, advance key technologies, and expand its team as the company continues to develop its proprietary field-reversed configuration (FRC) fusion system with the goal of commercialization.
Founded in March 2024 under the auspices of the University of Science and Technology of China, Xeonova builds on a fusion concept first proposed more than a decade ago by professor Sun Xuan. Sun developed the advanced field-reversed mirror (FRM) pathway, and since 2013, his research group has continued refining and validating the approach, with findings published in multiple peer-reviewed journals.
Backed by over a decade of R&amp;D and an established research base, Xeonova has built advantages in technological maturity, team capability, and execution. Sun, who has spent more than 20 years in fusion science, has worked across fusion laboratories, universities, and private companies in both China and the US. The team includes researchers from institutions such as the University of Science and Technology of China and Princeton University, bringing expertise across experimental physics and applied engineering.
According to Sun, the FRM configuration benefits from a linear reactor design that requires lower investment, offers structural simplicity, and allows easier maintenance and flexible control. These traits, he said, make it a potentially efficient, stable, and cost-effective route to commercial fusion energy.
As the global energy transition accelerates, controlled nuclear fusion is increasingly viewed as a cornerstone of the clean energy future, given its high energy density, abundant fuel supply, and intrinsic safety. A report by Keytone Ventures projects that the global fusion market will grow from USD 296.4 billion in 2022 to USD 395.1 billion by 2027.
Surging computing demands from <a href="/industry/ai">artificial intelligence</a> have also intensified attention on energy consumption, especially at data centers, positioning fusion power as a strategic priority for major tech firms. Microsoft has invested in Helion Energy and signed a 50-megawatt power purchase agreement, while Google has partnered with Commonwealth Fusion Systems on a 200-megawatt project. Policy momentum is also building: China’s 15th five-year plan designates nuclear fusion as a national priority, while the US and Germany have both launched dedicated programs and funding initiatives.
Multiple fusion approaches are now advancing in parallel. While tokamak-based systems remain dominant, they face persistent engineering challenges, including high construction costs, complex designs, and density limitations. The FRM configuration offers an alternative that’s simpler, cheaper, faster to iterate, and more adaptable. Progress across the field has also been rapid: Helion Energy has built seven generations of its reactor and aims to achieve net energy gain this year, with plans to deliver 50 MW of power by 2028. Meanwhile, TAE Technologies targets commissioning a hydrogen-boron fusion plant around 2030.
Xeonova expects its reactor facility to be completed and achieve first plasma discharge by year’s end, marking a key proof of engineering feasibility for its advanced FRM technology. The company said the milestone will lay the foundation for higher-performance reactors and advance China’s linear fusion pathway.
In parallel, Xeonova is exploring early commercial applications, including supplying diagnostic and heating components to universities and research institutes, as well as pursuing use cases such as neutron sources and isotope production. The company said these initiatives will help establish a foundation for near-term commercial opportunities in the fusion energy ecosystem.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3547106637754243">article</a> was written by Yuan Silai for 36Kr.

Meet Xeonova, an Ant-backed startup powering China’s nuclear fusion ambitions

Energy

… company is investing in proprietary IPs. In the first half of 2025, it acquired a 51% stake in HiToy for RMB 5.1 million (USD 714,000), gaining access to IPs such as Nommi, Honey, and MayMei. Ahead of its filing, Top Toy also acquired IPs including Ninimo.
Miniso founder Ye Guofu recently emphasized IP ownership as a strategic priority, saying Miniso will expand its IP development efforts and evolve from a retail company into a cultural and creative enterprise.
“Top Toy needs to convince Hong Kong investors that its IPs can generate value beyond its retail network,” said Gao Chengyuan, CEO of marketing firm Tiaoyuan. “At this stage, the company’s valuation can only lean toward high-turnover retail, with a price-to-earnings ratio of 25–30 times.” Developing strong proprietary IPs, he added, will be essential to securing a higher valuation.
whereRelatedArticle
<h3>The hard road to independence</h3>
If IP strength defines Top Toy’s competitiveness, its channel independence from Miniso may determine its market valuation.
As of the latest reporting period, Top Toy operated 38 directly owned stores, 250 franchise stores, five agency stores, and worked with 53 distributors. Revenue from distributors accounts for more than half of total sales.
Under the franchise model, partners operate stores and pay consulting fees to Top Toy while sharing revenue. In the first half of 2025, franchisees contributed 25% of total revenue. Distributors, meanwhile, purchase inventory from Top Toy and resell it to consumers. Miniso remains its largest distributor, accounting for RMB 923 million (USD 129.2 million) in 2024 and RMB 619 million (USD 86.7 million) in the first half of 2025, or 48.3% and 45.5% of total revenue, respectively.
Leveraging Miniso’s retail network has accelerated market penetration but also constrained Top Toy’s ability to establish an independent brand identity. Immersive flagship stores are key for designer toy brands to create emotional engagement, while Miniso’s variety store format dilutes that experience.
Heavy reliance on distributors also delays market feedback because consumer trends may take longer to reach the company. Directly operated stores, by contrast, offer real-time insights into Gen Z preferences and enable quicker product iterations.
To strengthen its brand, Top Toy plans to reduce dependence on distributors. Management expects its domestic store count to reach 380 to 400 by year’s end and aims to expand overseas to 100 countries with 1,000 stores within five years.
That strategy could influence valuation. “Excessive dependence on Miniso’s channels and franchise model may cap Top Toy’s long-term growth potential,” said Wang Peng, associate researcher at the Beijing Academy of Social Sciences. “Unless it proves its channels can operate independently, the Hong Kong market is likely to apply a valuation discount.”
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3547065866186887">article</a> was written by Chen Sizhu for 36Kr.

Top Toy’s Hong Kong IPO puts its IP strategy to the test

Miniso

Retail

Consumer Economy

Hong Kong

… costs wherever possible, including labor.
Ne Zha 2’s end credits list over 100 animation suppliers, an impressive sight that also exposes structural weaknesses. The long roster reflects a fragmented industry where top studios dominate while smaller ones remain stuck in low-value production work.
“There are far too many junior animators entering the market every year, but not enough senior ones,” an animator told 36Kr. “China’s animation films rely on sprawling supplier networks, while Western studios use project-based teams. The former gives big studios more bargaining power but doesn’t foster balanced growth.”
With its strong content pipeline and distribution network, Enlight Media already holds some of the best cards in China’s film industry. What it needs now is patience from the market—perhaps the rarest commodity in today’s climate.
Enlight was ahead of its time a decade ago. Today, it seems the times still haven’t caught up. As the company continues to bask in the glow of Ne Zha 2, its real turning point may come when Coloroom Pictures becomes the halo itself.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3545183355334535">article</a> was written by Song Wanxin for 36Kr.

Enlight Media caps record year, but questions linger after Ne Zha 2 boom

Entertainment

Creative

Content

Wondershare Technology, a global leader in creativity and productivity solutions, has announced the launch of <a href="https://filmora.wondershare.com/upgrade/15-free-download.html">Filmora V15</a>, marking a step toward democratizing artificial intelligence-powered video creation. Under the slogan “Edit as an Expert with Filmora AI,” the new version integrates AI across each stage of the editing process, redefining how creators produce and refine content and ushering in a new era of intelligent creativity.
Positioned as a next-generation all-in-one platform for professional video creation, Filmora V15 focuses on intelligent content generation and creative workflow management, connecting every step from idea conception to final output. The update also introduces a range of advanced features designed to simplify complex editing tasks and enhance creative control. Together, these upgrades aim to make workflows smarter, editing more efficient, and professional features more intuitive, empowering creators of all levels to produce high-quality content with greater speed and ease in the AI era.
<h3>AI at the core of creation</h3>
Filmora V15 introduces a new level of AI integration, embedding intelligent capabilities throughout the creative process. Rather than functioning as standalone tools, Filmora’s AI features are woven into the editing workflow, making AI a natural extension of creativity. As the first software to offer a complete end-to-end video creation experience—from AI content generation and extension to precise editing—Filmora V15 represents Wondershare’s bid to set a new benchmark for intelligent video production.
In Filmora V15, AI operates as a continuous production flow that connects every stage of creation, from ideation and generation to refinement and output. On the editing timeline, users can generate new scenes using “AI Video Generation,” remove distractions or backgrounds with “Smart Cutout,” and stretch footage using “AI Extend.” Combined with “Dynamic Captions,” “Voice Clone,” and “Text-to-Speech,” Filmora enables a seamless workflow that transforms talking-head videos into polished, ready-to-share content.
Among the most significant upgrades, “AI Extend” allows creators to intelligently continue clips forward or backward, filling missing footage with consistent visuals and smooth transitions. This solves common issues such as short clip duration or abrupt cuts. Combined with features including “Smart Cutout,” “AI Object Remover,” and “AI Video Enhance,” Filmora turns AI from a supportive tool into a creative partner that actively helps users produce professional videos faster and more efficiently.
<h3>Smarter tools for efficient editing</h3>
Beyond AI, Filmora V15 expands creative possibilities while simplifying advanced editing. New tools such as the “Pen Tool” let users draw motion paths or shapes directly on the timeline and apply dynamic stroke effects without third-party plugins. “Animated Charts” also allow users to bring data storytelling to life, transforming imported data into visually engaging bar and line animations with just a few clicks.
Additional updates include an expanded resource library with “Dynamic Captions,” “Text Animations,” and “Video Animations,” as well as cinematic filters, localized fonts, and trending visual assets. Together, these additions create an accessible ecosystem for diverse storytelling needs.
<h3>Simplified professional efficiency</h3>
Filmora V15 further improves professional editing efficiency with upgraded multi-sequence and project management capabilities. The new “Dual-Timeline Editing” feature enables editors to manage multiple compound clips and switch between timelines easily. “Subprojects” also allows users to import several project files as compound clips, simplifying complex workflows and enabling efficient content reuse.
Other enhancements such as “Source and Timeline Preview Editing” and “Multi-Track Video” boost decision-making and layering flexibility. Collectively, these updates bring Filmora closer to the sophistication of large post-production suites while maintaining the simplicity that creators appreciate.
<h3>Empowering global creators in the AI era</h3>
The launch of Filmora V15 comes amid the rapid adoption of AI across the global creator economy. According to <a href="https://market.us/report/creator-economy-market/">Market.us</a>, the market was valued at USD 149.4 billion in 2024 and is projected to grow at a compound annual rate of 21.8%, exceeding USD 1 trillion by 2034. From automated scene generation to real-time editing, AI is reshaping not only how videos are made but also who can make them.
“We’re seeing AI evolve from an assistive feature into a true creative partner,” said Christy Wu, vice president at Wondershare. “With Filmora V15, we’re moving toward intelligent collaboration between humans and machines, enabling creators to focus on ideas rather than manual processes.”
<h3>About Wondershare</h3>
Founded in 2003, Wondershare is a globally recognized software company known for its innovative solutions in creativity and productivity. Driven by its mission “Creativity Simplified,” Wondershare offers a range of tools, including Filmora and SelfyzAI for video editing; PDFelement for document management; and EdrawMax and EdrawMind for diagramming.
Citing over two billion cumulative active users across its product portfolio and availability in over 200 countries and regions, Wondershare strives to empower the next generation of creators with intuitive software and trendy creative resources, continually expanding the possibilities of creativity worldwide.
This press release was published in partnership with <a href="http://wondershare.com/">Wondershare</a>.

Wondershare launches Filmora V15 with deep AI integration to transform video editing

… growth phase. Over the next 10–20 years, he believes the global robotics market could reach USD 20 trillion, roughly twice the size of the automotive sector.
“I don’t know how many robots we’ll be selling ten years from now,” he said. “But I’m certain it’ll be more than cars.”
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3546815605567362">article</a> was written by Qiu Xiaofen for 36Kr.

Was there a person inside Xpeng’s robot? CEO debunks rumors onstage to viral disbelief

Xpeng Motors

Embodied Intelligence

… to help artists bring their ideas to life and position Aipy Studio as a global leader in IP operations.
Its talent pool is also growing rapidly. On the design side, Aipy Studio is actively recruiting from art colleges and has already received hundreds of submissions with encouraging early results.
Its partnership with the Sichuan Fine Arts Institute includes three initiatives:
<ul>
<li aria-level="1">Talent development through co-created courses and workshops that integrate technology with music, design, and innovation.</li>
<li aria-level="1">Research collaboration to establish a joint center for design standards and an AI aesthetics library.</li>
<li aria-level="1">Career pathways via internships and employment channels connecting students to real-world projects.</li>
</ul>
These efforts reflect a broader ambition to merge technology and creativity. With AI as its foundation, Aipy Studio seeks to move the IP industry from experience-driven to intelligence-driven growth.
This approach also feeds back into iFlytek’s hardware business. Lu shared an example: when the company launched an AI-powered learning tablet featuring a Ne Zha 2 limited-edition design, it quickly captured children’s attention, she said.
“The fusion of IP and hardware is powerful, enabling the connection of products with user communities on an emotional level. Mastering IP operations can create great value for our core business,” she said.
Ultimately, Aipy Studio aims to become a catalyst for the future of Chinese IP: one where AI democratizes creativity, reduces production costs, and empowers everyone to create and express.
KrASIA Connection features translated and adapted content that was originally published by 36Kr. This <a href="https://36kr.com/p/3542801353142407">article</a> was written by Xiao Xi for 36Kr.

iFlytek enters IP market with Aipy Studio, betting on AI to fuel its boom

iFlytek

Creator Economy