Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Zhipu AI

AI overview

Sign in and the AI will write an overview from our coverage.

Headlines · 75

  1. AI reshapes China’s 2026 Hurun Billionaires List

    The 2026 Hurun Billionaires List estimates Zhang Yiming’s wealth at RMB670 billion, returning him to first place, while DeepSeek founder Liang Wenfeng enters at seventh with an estimated RMB280 billion. Reports from AIX财经, 36氪 and 钛媒体 say more than 160 listed entrepreneurs now have fortunes linked to AI, spanning foundation models, robotics and computing infrastructure. These are valuation-based estimates rather than liquid personal wealth, and ByteDance’s established Douyin and TikTok businesses remain major sources of revenue and cash flow.

    虎嗅 AI · 🔥 0
  2. Beijing Qujing Technology raises over RMB 1.3 billion in Series B

    Beijing Qujing Technology has completed a Series B financing round exceeding RMB 1.3 billion, according to the supplied reports. Zhejiang Financial Innovation, Xiaoshan Capital, ICBC Investment, Zhongguancun Science City, Xinglian Capital and Qianjiang Century City jointly led the round, while SenseTime, Wangsu Technology, CITIC Juxin and others participated. Separately, Qujing, Zhipu AI and Moore Threads jointly released a domestic heterogeneous-inference solution for high-quality AI Token production.

    36氪 人工智能 · 🔥 0
  3. 2026 Hurun Rich List: DeepSeek Founder Liang Wenfeng Debuts at No. 7

    Hurun Research Institute's 2026 Hurun Rich List includes 1,529 entrepreneurs with wealth of at least RMB 5 billion, with combined wealth of RMB 32 trillion. DeepSeek founder Liang Wenfeng debuted in seventh place with an estimated RMB 280 billion, while ByteDance founder Zhang Yiming ranked first with RMB 670 billion. Zhang's wealth increased by RMB 200 billion from the previous year, and one report attributed the increase to AI business progress lifting ByteDance's valuation. IT之家 and cnBeta described Zhang's first-place result as his second time becoming China's richest person, while another report described it as consecutive retention of the title. cnBeta also reported that more than 160 listed entrepreneurs had wealth tied to AI.

    IT之家 AI · 🔥 0
  4. AI Token Sales Reach E-Commerce as API Intermediaries Face Pressure

    Tech星球 reports that Alibaba Cloud, DeepSeek, Zhipu AI, MiniMax and Kimi have begun offering AI subscriptions and Token top-ups through e-commerce platforms. Tmall was also discussing bundles for students, families and solo creators, but the report did not say those bundles had launched. The analysis describes weak profitability among API intermediaries, risks associated with unofficial endpoints and account sales, and regulatory enforcement against services that lacked required assessments or faced content-safety issues. It also identifies emerging models including authorized distribution, model calls bundled with cloud services, telecom Token packages and automatic model routing. Reported financial figures and other claims were not independently verified in the supplied material.

    虎嗅 AI · 🔥 0
  5. AI Shopping Assistants Enter Double 11 Transaction Flows

    TMTPost reports that Doubao can direct shoppers to Douyin livestreams and apply channel coupons, while Qwen’s Double 11 shopping page supports product selection, ordering and payment in its app. A media comparison found one kitchen tissue costing 27.9 yuan through Douyin e-commerce but 17.43 yuan through Doubao after subsidies and coupons, although this may redirect an existing order rather than create new demand. TMTPost also says merchants are already paying for office AI, with Qwen Office reporting that 20% of its paid users come from e-commerce.

    钛媒体 · 🔥 0
  6. GLM-Omni Research Tests Knowledge Transfer From Images to Language

    A joint Tsinghua University and Zhipu AI paper disclosed the previously unreleased internal model GLM-Omni, with Zhipu chief scientist Tang Jie as corresponding author. In the MagicWorld experiments, GLM-Omni achieved 54% average visual-generation accuracy but only 14.57% on text answers; interleaved image-text alignment raised text accuracy to 26.86%, while Visual Chain-of-Thought further raised it to 41.14%. The paper did not establish which product version, if any, corresponds to GLM-Omni.

    雷峰网 AI · 🔥 0
  7. Zhipu Compensates Coding Plan Users and Teases Next GLM Model

    Zhipu reportedly sent affected GLM Coding Plan users three quota-reset cards after some encountered model degradation, according to 36Kr, which republished QbitAI. A compensation notice also promised another quota reset after a new model launches. Reports circulated about a possible GLM-5.5 model, including unverified estimates of 2.5–3.5 trillion total parameters and the possible use of a recurrent Transformer, but Zhipu has not confirmed its name, release date, specifications, or architecture.

    36氪 人工智能 · 🔥 0
  8. Zhipu Wins RMB 10.58 Million Xiamen Healthcare AI Project

    Zhipu AI won the bid for Xiamen Health Medical Big Data Center’s “Artificial Intelligence Medical Large-Model Foundation and Application Engineering Phase I” project, with an award amount of RMB 10.58 million against a budget and maximum price of RMB 10.8 million. The procurement covers a medical large-model platform, data-security coordination and federated learning, intelligent mental-health services, data asset management, and security operations. Platform functions such as model management, service invocation, training evaluation, runtime monitoring, and compute scheduling are stated procurement requirements, not independently verified capabilities.

    智东西 · 🔥 0
  9. MiniMax Shares Fall Below IPO Price After Post-Listing Rally

    MiniMax shares fell as low as HK$159.3 during intraday trading on October 22, below the HK$165 IPO offering price, according to IT之家. The stock had reached HK$1,330 in March, more than seven times its offering price, before the post-listing rally reversed. 虎嗅 AI commentary attributes the pressure to competition, commercialization challenges and a lower-margin shift toward business customers.

    IT之家 AI · 🔥 0
  10. China's Big Tech races to hire Forward Deployed Engineers, with pay packages topping 1 million yuan

    A China Entrepreneur feature, republished by Huxiu AI and 36Kr, reports that ByteDance, Alibaba, Tencent and Kingsoft have spent the past half year urgently recruiting Forward Deployed Engineers (FDEs), a role it says Palantir popularized around on-site delivery that embeds standardized products into customer systems. Citing listings on BOSS Zhipin and Liepin, the report says Doubao posted an FDE package of 1.05 million yuan a year, Alibaba Cloud's large-model FDE role pays 20,000–55,000 yuan a month over 16 months, and Moonshot AI's Kimi listed a 1.09 million yuan annual package; a person close to ByteDance said some ByteDance FDEs placed at client sites earn up to 10,000 yuan a day, a claim not confirmed by the company. The article also relays industry skepticism, with investors and practitioners saying much FDE work amounts to training, that Chinese AI-to-B delivery remains project-based, and that one investor estimated OpenAI's FDE team at roughly 50 people by the end of 2025 while Chinese vendors build delivery teams hundreds strong.

    虎嗅 AI · 🔥 0
  11. Kimi K3 reportedly pulled from sale; developer tests its code-editing on a 3D pool game

    A Juejin author says he was testing Kimi K3 one day and the next day "heard" it had been taken off sale, adding that GLM 5.2 has stayed in a purchase-limited state since becoming popular. Neither claim is backed by vendor confirmation, and the author himself frames the K3 halt as hearsay. The bulk of the post is a hands-on test: over 12 rounds of dialogue he had K3 build a networked 3D billiards game using Three.js and hand-written physics; the first-round output had serious modeling problems (no pockets cut into the table) and erratic camera control, and only after many revisions did it reach a playable state. He rates K3 as broadly top-tier among domestic models with good scene modeling and summary documentation, but criticizes the poor first draft, long iteration, heavy token use and weak autonomous planning.

    掘金 · 人工智能 · 🔥 0
  12. NetEase LobsterAI report: China office agent DAU up 83.1% in September

    NetEase LobsterAI published the September edition of its China Office Agent User Behavior Report, saying daily active users of office agents rose 83.1% month over month in September, faster than the 19.8%, 25.1% and 30.6% it lists for June, July and August; all figures come from the vendor's own report with no independent measurement disclosed. It says nearly 70% of users rely on a single model day to day while heavy users average 2.4 models, that combinations built on the DeepSeek Flash series fill all top five pairing slots, and that DeepSeek V4.1 Flash posted a 97.3% cache hit rate, leaving an effective cost of 4.6% of list price. The report also says Zhipu's GLM Token share jumped about 166.7% month over month on the back of a new model, that Doubao Seedream and Seedance took 57.2% and 54.4% of image and video generation tasks respectively, and that 77% of users assign tasks through WeChat, with single-task Token volume reaching a record 202,000.

    品玩 实时要闻 · 🔥 0
  13. Runjian signs $200M cross-border AI token contract

    Citing a report by Alpha Works Institute republished by 36Kr and CYZONE, Chinese listed company Runjian disclosed a $200 million cross-border AI services contract in which an overseas client calls large-model inference through APIs while compute stays in domestic smart computing centers. The one-year deal runs from October 19, 2026 to October 18, 2027, is billed post-paid by usage with monthly settlement, and includes a minimum-purchase guarantee requiring the client to make up any shortfall below $200 million in token service fees; the report says the amount equals roughly 35.6% of Runjian's first-half 2026 revenue and involves Wuxiang Yungu as the service provider. The report also notes the company's computing-network segment posted a 6.97% gross margin, down 2.86 points year on year, and negative operating cash flow of about 2.876 billion yuan, and cites the filing's warning that the token business demands heavy capital, technology and customer development and that large-scale commercialization remains uncertain.

    36氪 人工智能 · 🔥 0
  14. Tmall's Double 11 debuts 'Super Flagship' AI brand agent; AI-tagged apparel lifts conversion nearly 20 points

    A TMTPost-published article says Tmall's 2026 Double 11 began its first round on the evening of Oct 15, with full-category sales starting Oct 20 and running to Nov 13. Tmall introduced "Super Flagship," which it describes as upgrading AI from a marketing tool into a "digital employee" spanning brand operations, generating outfit recommendations from browsing and purchase history plus weather, occasion and fabric trends. The article also says apparel carrying AI-translated quality labels saw order conversion rise nearly 20 percentage points and that such high-quality goods carry average order values more than double ordinary items — figures attributed to the platform. It adds that model makers Zhipu and MiniMax opened Tmall flagship stores before the event to sell tokens.

    钛媒体 · 🔥 0
  15. Trillium founder: RSI is a narrative, distillation can't be stopped

    In an a16z podcast interview with Lisha Li, Trillium founder Nathan Lambert argued that the recursive self-improvement narrative overstates the prospect of a sudden AI takeoff and said his bigger concern is AI labs' poorly secured infrastructure. He framed model progress as a function of compute, data and talent, described Chinese labs as compute-constrained but highly efficient, and said distillation cannot be stopped through terms of service or legislation. Lambert also presented open-weight releases as a pragmatic way for Chinese labs to enter the market and gain influence.

    36氪 人工智能 · 🔥 0
  16. Hong Kong AI model stocks slide, with Zhipu and MiniMax each down more than 10%

    Hong Kong-listed large-model stocks fell on October 15, with Zhipu and MiniMax each dropping more than 10%, according to 36Kr Express. At the close, the Hang Seng Index was down 1.03% and the Hang Seng Tech Index 2.72%, led by semiconductors: Montage Technology fell over 11% and NOVOSENSE over 8%, while hardware names such as YOFC dropped more than 13%. The closing report put Lenovo Group down over 14%, MiniMax over 12% and Zhipu over 10%, while consumer and banking shares gained, with ICBC, China Construction Bank and JD Health up more than 1%, and southbound flows recording net selling of HK$5.08 billion.

    36氪快讯 · 🔥 0
  17. MiniMax advances A-share IPO with second CITIC Securities tutoring report

    CITIC Securities published the second progress report on its pre-listing tutoring work for MiniMax's planned A-share initial public offering, covering a tutoring period from July 10 to October 13, 2026. During that period, the intermediaries said they completed full due diligence, held group tutoring sessions and kept regular communication with the company, while pressing MiniMax to clarify its business goals, future development plans and the use of proceeds. CITIC said MiniMax has set up internal governance structures and rules but that execution still has room for improvement, and the next phase will see the tutoring team work with Commerce & Finance Law Offices and Ernst & Young on rectification and preparation for the tutoring acceptance review. Separately, a 36Kr brief citing Shanghai Securities News said Zhipu, MINIMAX and other Hong Kong-listed companies are progressing with plans to return to A-share listing, which industry figures described as rational capital operations based on corporate structure, sci-tech attributes and profit cycles.

    AIbase AI新闻 · 🔥 0
  18. As token prices fall, AI model vendors rewrite the pricing yardstick

    An analysis republished by 36Kr (originally from 窄播AI) says that as per-million-token prices keep falling, model vendors are separately pricing speed, resource priority and budget certainty. It cites OpenAI's July moves — cutting GPT-5.6 Luna API input and output prices by 80% and mid-tier Terra by 20% — while selling a Fast Mode for GPT-5.6 Sol at twice the standard price for up to 2.5x the output speed, and an early-October update that raised GPT-6 Astra and GPT-6.1 Sol speeds by about 50%. OpenAI CFO Sarah Friar is cited proposing "Useful Intelligence per Dollar," arguing enterprises should count task completion, time spent, retries and human supervision rather than token list prices. The piece cites Ramp data showing effective price per million tokens for its US business sample fell from $1.15 to $0.68 (41%) between March and September 2026 while monthly token spending grew 20x from June 2025 to June 2026, and SemiAnalysis subscription-quota tests estimating that, assuming a 92% API gross margin and fully used quotas, exhausting Opus 5.5 quota implies roughly -369% subscription gross margin versus about 1% for Fable 5.1.

    36氪 人工智能 · 🔥 0
  19. Silicon Valley's AI subsidy war tests the cost edge of Chinese models

    Tech星球, republished by 36Kr and TMTPost, surveys how OpenAI, Anthropic and Google are lowering entry costs through API pricing tiers, model segmentation and free subscription credits. It cites a SemiAnalysis test that converted subscription plans into API value using its September agent usage mix, about 96.6% of which was cache reads. SemiAnalysis found a $200 Claude Max 20x plan could yield roughly $12,529 of Sonnet 5.5 usage, about 62.6 times the fee, versus about 23.4x for MiniMax Token Ultra, 11.6x for Zhipu GLM Coding Max and 6.7x for Kimi Code Ultra. The author argues a low per-token price does not guarantee a low task cost, and cites developers who say they burned quota quickly on Chinese plans.

    36氪 人工智能 · 🔥 0
  20. Epoch AI: Six Chinese AI firms' revenue is about 10% of OpenAI and Anthropic combined

    Citing a new Epoch AI report, 36Kr says the AI-related revenue of Alibaba, ByteDance, Zhipu, Moonshot AI, DeepSeek and MiniMax totaled about 10% of OpenAI and Anthropic's combined revenue as of September 2026. Epoch AI estimated annualized revenue of about $65 billion for Anthropic and $40 billion for OpenAI, versus roughly $10 billion combined for the six Chinese firms; ByteDance's and Alibaba's figures mainly reflect model-as-a-service revenue. The report outlines five monetization paths and notes that open-weight diffusion pressures API pricing while consumer paid conversion and gross margins remain low.

    36氪 人工智能 · 🔥 0
  21. Moonshot AI reported at $50B valuation as it moves toward a 2027 Hong Kong IPO

    Citing people familiar with the matter, Bloomberg reported on October 6 that Moonshot AI has closed a final funding round at a valuation of about $50 billion and plans to list in Hong Kong in the first quarter of 2027, according to articles republished by 36Kr and TMTPost from Bohu Finance. That is about a month after Moonshot was reported to have filed a confidential A1 listing application with the Hong Kong Stock Exchange. The articles say K3, released in July, topped the Hugging Face trending list within 30 minutes, and that annualized recurring revenue rose from $100 million in March to $1 billion in August, with Bloomberg's sources pointing to $2 billion by year-end. They also list September commercialization moves, including the "Moon Landing" partner program, the full rollout of K2.8 Preview, K3 on Amazon Bedrock, and the release of Kimi Code Desktop and a financial-industry solution. The funding and IPO details rest on unnamed sources, have not been confirmed by the company, and contrast with founder Yang Zhilin's late-2025 internal letter saying a listing was not an immediate priority.

    36氪 人工智能 · 🔥 0
  22. Anthropic says anti-distillation defenses cut Claude reasoning theft 90-99%

    Anthropic said on Monday, October 13, that it has rolled out new defenses meant to stop competitors from replicating Claude's capabilities through distillation, and that the volume of reasoning tokens extracted on its own platform has fallen by roughly 90% to 99%, to one-tenth to one-hundredth of previous levels. The measures include preserved thinking in Fable 5.1 and Opus 5.5, which stops protected reasoning from being reused once earlier context is altered, a two-stage classifier that blocks requests trying to elicit full reasoning, and rate limits or bans for accounts that keep probing; Anthropic says the system is roughly as robust as its defenses against Claude helping with bioweapons, with a false-positive rate below 0.1%. Anthropic says it has banned tens of thousands of related accounts and named Zhipu AI as having tried to distill from Claude while developing GLM-5.3, while The Decoder also reports Moonshot was named for routing queries through Claude; the allegations are Anthropic's. Anthropic acknowledges legitimate requests can still be misclassified and that blocked requests have been billed since September 24.

    IT之家 AI · 🔥 0
  23. Peter Thiel: a model-era Google may never emerge

    Peter Thiel, an investor in OpenAI and Anthropic, received the Axel Springer Award in Berlin in the last week of September and discussed AI with Mathias Döpfner. He argued that the model layer looks more like interchangeable intelligence than a unique gateway: switching costs are low, product differentiation lasts only months, and much of the value may reach users as consumer surplus; he framed the view as an "I wonder." This commentary by Jinduan, republished by 36Kr and TMTPost, adds that chips, power, distribution gateways and irreplaceable workflow-context layers may still capture strong profits.

    36氪 人工智能 · 🔥 0
  24. Jensen Huang: Chinese open models like Kimi are 'good enough,' backs open source

    Speaking at a forum hosted by DBS Bank, NVIDIA CEO Jensen Huang said Chinese AI models including DeepSeek, Kimi and GLM are "already good enough." He described himself as a firm supporter of open source, calling it the best way to drive efficiency gains, and said the market will decide which models best fit different use cases. Huang also rejected the framing of AI competition as "a race with a finish line," saying both the US and China will be long-term players, and argued that cheaper, more open models will let more people and companies benefit from AI, increasing demand for NVIDIA chips, data centers and compute. The report says he argued the market first misread DeepSeek's impact and that Wall Street is now misreading Kimi's, and that quantum computing could reach mainstream adoption within three to five years.

    cnBeta · 🔥 0
  25. Wang Yuquan: AI is no bubble, but 2029 is the risk window

    Technology investor Wang Yuquan of Haiyin Capital argues in a conversation that the current AI boom is not a bubble but a decade-plus technology shift, with the real risk window around 2029, when application paradigms in AI software, AI coding and autonomous driving become clear while revenue lags behind infrastructure spending. He expects 24/7 trading, quantitative strategies and stablecoin growth to amplify market swings, hurting memory makers such as SK Hynix and South Korea hardest, while leaders like Nvidia are undervalued. He also calls embodied AI and large language models overhyped, urges Chinese labs to close their models for safety and revenue reasons, and predicts a 70-30 split favoring China outside the US.

    虎嗅 AI · 🔥 0
  26. Guanghe Ventures' Zhu Jia on investing up the AI stack from terminals to chips

    In an interview with Cyzone, Guanghe Ventures partner Zhu Jia described the firm's hard-tech method of starting from downstream end products and working upstream through the supply chain, citing investments in MetaX, Unitree Robotics, Zhipu AI, Hesai and Innolight. The article says the firm recorded 10 portfolio IPOs in the past 18 months and in April closed more than RMB 5 billion across two dual-currency funds. Zhu summed up his approach as backing the number-one player in any large细分 market, and said the team had already made early bets such as X Square Robot before embodied-AI valuations heated up; he also discussed slower-developing portfolio company Yunmai Xinlian.

    创业邦 科技 · 🔥 0
  27. Zhipu AI grey-tests GLM-5.3 HighSpeed in ZCode at over 490 tokens/s

    Zhipu AI has quietly started a limited grey test of its GLM-5.3 HighSpeed model inside its AI coding tool ZCode, according to 智东西 and 36氪 人工智能, which republished the same report. AI coding blogger Leo said his grey-test instance exceeded 490 output tokens per second, while the AI Ping platform cited in the report measured the standard GLM-5.3 API at 70.84 tokens/s and DeepSeek V4.1 Flash's official API at 121.34 tokens/s. The variant is randomly opened to ZCode subscribers with no application entry point, and it had already appeared on the open-source coding agent Pi harness site about a month earlier with pricing undisclosed. Separately, a Reddit user reported that the open-source Project Maya runtime raised their local GLM-5.3-Flash throughput from about 10 to 30 tokens/s.

    智东西 · 🔥 0
  28. At 28, Yao Shunyu sits beside Ma Huateng as China's AI talent war heats up

    An analysis piece notes that when Guangdong party secretary Huang Kunming visited Tencent's Shenzhen headquarters on Oct. 8, 28-year-old Yao Shunyu sat at Ma Huateng's right hand. A Tsinghua Yao Class graduate and Princeton PhD who joined OpenAI in 2024 to work on Operator and DeepResearch, Yao joined Tencent in December 2025 as chief AI scientist in the CEO/president's office and now runs its foundation model, AI infra and AI data units covering the full Hunyuan stack. The article also reports that DeepSeek raised more than 50 billion yuan in a first round, opened a second round of the same size and is preparing an IPO, after at least five core researchers left for Xiaomi, ByteDance, Tencent, Baidu and an autonomous-driving startup.

    虎嗅 AI · 🔥 0
  29. Virtual quantum lab stress test: 17 AI agents, pass rates from 78% to 3%

    A joint team from the National University of Singapore, Nanyang Technological University and GaugeForge released Quantum-Harbor, a simulated quantum engineering lab, plus QIQCBench, a set of 49 real-world tasks spanning measurement and characterization, control, error correction, compilation, simulation and sensing. Seventeen frontier AI agents were dropped into the physics-driven environment with instrument APIs and hidden hardware parameters, then graded under an "evidence-bound" verified-autonomy standard that checks whether conclusions are supported by data the agent actually collected. Pass rates ranged from 78% at the top down to 3%, 937 of 1,510 failures came from exhausting budget or looping, and the many-body physics category averaged only 15%.

    36氪 人工智能 · 🔥 0
  30. GLM-5.3 joins Amazon Bedrock, showing two Chinese model go-to-market paths

    An analysis published via 36Kr's Muhe Business Finance says Zhipu's GLM-5.3 went live on Amazon Bedrock during China's October National Day holiday under a usage-based revenue-share deal; Zhipu's Hong Kong shares reportedly rose more than 7% that day, and Goldman Sachs upgraded the stock to buy with a HK$1,560 target, raising its end-2026 ARR estimate for Zhipu from $2.7bn to $3.2bn. The piece contrasts this with DeepSeek, which has been available on Bedrock since March 11, 2025 via Bedrock and Custom Model Import but is described as prioritising distribution and open weights over direct monetisation, a difference it ties to Zhipu's post-IPO revenue pressure versus DeepSeek's High-Flyer backing and its reported RMB 51bn first external round in June 2026.

    36氪 人工智能 · 🔥 0

← PreviousNext →

Experience and discussion from the community

Share my Zhipu AI experienceAsk about Zhipu AI

Nobody has shared their experience with Zhipu AI yet.