However, in testing,there is still a gap compared to SOTA products.<\/strong>The introduction and filtering of high-quality sources still need improvement. The operational logic of web search appears somewhat immature. The presentation of results lacks depth and is prone to omissions.<\/p>\n In actual use,it can be used as an important reference, but it is not recommended to use it directly without verification.<\/strong>\u3002<\/p>\n Below is a practical case study comparing it with OpenAI’s DeepResearch:<\/p>\n Just recently, Zhihu released its Q4 2024 financial report, achieving its first single-quarter comprehensive profit. We will use the analysis of Zhihu’s commercial development over its 14-year history as the theme for the research task.<\/p>\n #### Basic Information<\/p>\n Zhihu was founded in 2010 by former ‘Southern Metropolis Weekly’ journalist Zhou Yuan and others. The website name means ‘Do you know?’ It was initially positioned as a high-end Q&A community, using an invitation-only registration system to attract experts from fields such as the internet, investment, and academia to share professional knowledge. After its official launch in January 2011, it received angel investment from Lee Kai-fu’s Innovation Works and others, starting with a ‘small circle, high-quality content’ approach. In the first two years of its establishment, Zhihu maintained an invitation-only operation model, with only 200 seed users (including well-known industry figures like Lee Kai-fu and Wang Xing), laying the foundation for a high-quality discussion atmosphere in the community.<\/p>\n Ten years later, Zhihu stepped onto the capital market. On March 26, 2021, Zhihu was listed on the New York Stock Exchange under the ticker symbol ‘ZH’, with an offering price of $9.5 per ADS, raising approximately $522 million. At the time of its IPO, Zhihu’s valuation was about $5.3 billion. One year after listing, on April 22, 2022, Zhihu achieved a dual primary listing on the Main Board of the Hong Kong Stock Exchange (stock code: 2390), with an upper limit offering price of HK$51.8. Early investors such as Innovation Works and Qiming Venture Partners sold old shares to realize gains. It should be noted that Zhihu has maintained its listed status on both the US and Hong Kong stock exchanges without delisting or transferring. At the time of listing, Zhihu’s cumulative registered users had exceeded 200 million, and it was regarded as ‘China’s largest online Q&A community’.<\/p>\n #### Pre-IPO Development (2010\u20132020)<\/p>\n From its establishment to before its IPO, Zhihu went through stages such as community cold start, user explosion, and commercialization exploration, with significant business progress and strategy changes each year.<\/p>\n #### Post-IPO Development (2021\u20132025)<\/p>\n After entering the listed company stage, Zhihu faced the performance demands of the capital market. Its business focus gradually shifted from solely pursuing user growth to balancing growth with profitability. The following provides an overview of Zhihu’s post-IPO development trajectory by year\/quarter, including financial indicators, business strategies, market reactions, and organizational changes:<\/p>\n Chart: Zhihu’s Revenue and Net Profit Trends from 2019 to 2024 (Unit: 100 million RMB, negative values indicate losses). It can be seen that Zhihu’s revenue grew rapidly from 2019 to 2021, but net profit remained consistently negative; starting from 2022, the growth rate slowed, and the company deliberately controlled losses, achieving its first single-quarter profit in Q4 2024.<\/p>\n #### External Investments and Acquisitions<\/p>\n Zhihu conducted several investment\/acquisition actions highly related to its core business before and after its listing, mainly focusing on the vocational education sector. Below lists Zhihu’s important external investment and acquisition events:<\/p>\n Table: Overview of Zhihu’s External Investments\/Acquisitions in Vocational Education. Zhihu rarely makes investment moves outside its core business, with vocational education becoming an exception as a key focus area. Since 2021, it has consistently invested in training sectors like finance, study abroad, and teacher qualifications, demonstrating its “determined to win” stance in this field.<\/p>\n Investment Logic and Domain Evolution: Zhihu’s foray into the vocational education sector stems from the large market space of vocational education and its natural synergy with the Zhihu Q&A community. “Everyone’s problems ultimately are education problems” \u2013 Zhou Yuan’s thinking from years ago has now become Zhihu’s strategy. Zhihu noticed that the community is filled with Q&A related to postgraduate exams, civil service exams, vocational skills, etc., with users having strong learning and improvement needs. According to a Frost & Sullivan report, China’s non-degree vocational education and training market is expected to reach 110.2 billion yuan in 2026, with a compound annual growth rate of 21.4% from 2019\u20132026, making it an attractive “big cake”. However, this industry has long been fragmented and inconsistent, lacking a leading platform. Zhihu identified the alignment between market pain points and its own advantages: Zhihu has over 100 million knowledge-oriented users and abundant professional answerer resources. If it can convert community traffic into education services and increase industry information transparency, it has the opportunity to stand out among many vocational education players. Based on this, Zhihu began attempting to collaborate with external vocational education institutions to sell courses through joint operations as early as 2019, gradually accumulating experience. After 2021, Zhihu even began frequently investing in and taking stakes in education-focused companies, a stark contrast to Zhihu’s traditionally conservative investment style. In terms of target selection, Zhihu has expanded from finance and accounting (Pinzhi Education) to language exams (Papa Education) to civil service and teaching recruitment (Zhangshang Yuanbu), covering increasingly broader areas, almost encompassing the important stages of young people’s career growth. The investment steps have also progressed from minority stakes to direct acquisition and integration, reflecting Zhihu’s determination to build a complete vocational education business map. By the end of 2023, Zhihu had established a multi-directional curriculum system covering “postgraduate studies \u2013 study abroad exams \u2013 professional qualifications \u2013 civil service\/teacher recruitment”, consolidating its “second growth curve” through self-operation, cooperation, and acquisitions.<\/p>\n Synergy Analysis: Zhihu’s frequent moves in vocational education show significant synergy with its core business. First, traffic synergy: Zhihu has a large number of young users with learning needs. For example, topics like “postgraduate exams” and “national exams” have high attention on Zhihu (with view counts often reaching tens of millions for related questions). Directing this precise traffic to Zhihu’s self-operated or partnered courses can greatly reduce customer acquisition costs and increase conversion rates. Zhihu provides an in-app “Learning” section entrance, naturally connecting the community Q&A scenario to the course purchase scenario. Second, content synergy: The Zhihu community itself has accumulated a massive amount of high-quality answers, which can serve as supplements and selling points for educational content. For example, finance training can cite practical experience answers from excellent Zhihu answerers, and English training can incorporate insights shared by Zhihu study-abroad returnees. Zhihu can also invite KOL answerers from the platform to participate in course teaching or live broadcasts, linking community content with course products. Third, brand synergy: The Zhihu brand is closely associated with knowledge and trustworthiness, providing a good endorsement for its entry into the education field. Many users are willing to trust the quality of courses from “Zhihu”, a trust that pure training institutions lack. Finally, Zhihu’s AI technology investment can also feed back into its vocational education business \u2013 utilizing big data and models to provide students with services like intelligent practice question solving and job matching, enhancing the teaching experience. Of course, synergy does not mean without challenges. Integration risks emerged during the acquisition of Zhangshang Yuanbu; Zhihu needed to merge an external team’s culture and systems into its own, ensuring consistent service quality. Fortunately, so far, the product reputation of Zhihu Zhitang is relatively good, with integration proceeding smoothly (for example, “Yiqi Kao Jiaoshi” launched successfully on the Zhihu learning platform with positive user feedback). After successfully piloting several cooperations, Zhihu gradually developed a model, upgrading the joint operation model to a deeper co-building operation, and even directly acquiring quality targets to fully control course quality and service. In contrast, Zhihu’s past attempts in community e-commerce (such as the male-focused community “CHAO”) have quietly been taken offline due to insufficient synergy. This positive and negative comparison further highlights that Zhihu’s choice of the vocational education track is well-considered, having indeed found a monetization breakthrough with high alignment to its core business.<\/p>\n Success and Failure Cases: From the current progress, Zhihu’s investment in the vocational education field has shown initial results and can be considered a successful case at this stage. In the first three quarters of 2023, Zhihu’s vocational education business revenue reached 145 million yuan, with its share rising to 14%. The Zhihu Zhitang brand also gained quick recognition, launching multiple AI features after its independent operation in early 2024, gaining market attention upon debut. Of course, there have been some detours on this path. For example, in the early days when Zhihu jointly sold courses with third-party institutions, due to long user conversion paths and inconsistent service, the actual results were mediocre. Zhihu realized that merely acting as a traffic diversion platform was hard to generate word-of-mouth, so in 2023 it adjusted to deeply participate in operations and even self-develop courses. This meant increased investment, but greater control over user experience, leading to improved repurchase rates and reputation. Another potential risk is educational regulation: in recent years, policies on education and training have been volatile (such as the “double reduction” policy affecting the K12 sector). Although Zhihu focuses on adult vocational education, which is relatively less regulated, it still needs to pay attention to possible policy requirements for vocational education. However, overall, Zhihu’s choices in external investments and acquisitions have been very restrained and focused. All known investments currently serve the core knowledge payment business, with no cases of cross-border acquisitions leading to scattered management. This cautious strategy also reduces the possibility of major failures. In summary, Zhihu’s external investment layout has high synergy with its core business. Through successful integration of educational resources, it has not only opened up new revenue channels but also further strengthened Zhihu’s positioning as a “knowledge service platform”, forming a positive feedback loop for its overall business.<\/p>\n #### Key Strategic Turning Points<\/p>\n In the history of Zhihu’s development, there have been several milestone-like strategic turning points that have profoundly influenced the company’s medium to long-term trajectory. This article outlines 5-7 major turning points, explaining the background, reasons, and outcomes of each, and analyzing their long-term impact.<\/p>\n 1. 2013 Open Registration: From Elite Community to Mass Platform. During the first two years after its establishment, Zhihu maintained an invitation-only system. This small-scale operation ensured content quality but limited user growth. By early 2013, Zhihu had only about 400,000 registered users. The founding team realized that to build a business model, they needed a larger user base and traffic support. Meanwhile, competitors like Baidu Zhidao and other mass Q&A platforms had already accumulated massive users and content. If Zhihu continued to be closed, it might miss the window of opportunity. Against this backdrop, Zhihu abolished the invitation-only system in March 2013 and opened registration to the public. After opening registration, user numbers grew exponentially, exceeding 4 million in less than a year and continuing to climb in the following years. The result of this turning point was Zhihu’s transition from a niche platform to a mainstream one, rapidly ‘breaking the circle’ to become a well-known content community on the Chinese internet. The long-term impact is two-fold: on one hand, the massive influx of new users provided the foundation for Zhihu’s subsequent commercialization\u2014without the user explosion in 2013, there would be no advertising and membership revenue later. On the other hand, the change in user composition also impacted the early elite atmosphere. The community began to see entertainment-oriented and low-quality content, with old users lamenting that ‘Zhihu is no longer pure.’ Zhihu later had to introduce mechanisms like upvote\/downvote systems and level-based permissions to improve content filtering and adapt to a more mass-market community. Overall, the 2013 open registration was one of the most important turning points in Zhihu’s development: it laid the foundation for Zhihu to become China’s largest knowledge community and also meant that Zhihu had to face the governance challenges of a ‘large community.’<\/p>\n 2. 2016 Commercialization Kickoff: Exploring Profit Models, Shifting from Pure Community Operations to ‘Community + Business’ Dual-Track Development. After six years of establishment, despite having tens of millions of users and a good reputation, Zhihu had not found a clear profit model, which once raised doubts among investors. Around 2016 coincided with China’s ‘Year One of Knowledge Payment,’ with the rise of knowledge service platforms like Dedao and Ximalaya. Zhihu’s management realized that continuing to ignore commercialization would miss opportunities and even endanger the company’s survival (after all, financing always requires returns). In 2016, Zhihu launched a series of commercialization attempts, marking a major strategic shift: first, launching advertising services, experimenting with in-feed native ads and branded columns, and beginning to monetize traffic. Although Zhihu’s advertising scale was not large at the time, this meant Zhihu moved from idealism to pragmatism, starting to bear revenue pressure. Second, launching paid products ‘Zhihu Value’ and ‘Zhihu Live.’ ‘Zhihu Value’ as an attempt at paid Q&A allowed users to directly pay for high-quality answers; ‘Zhihu Live’ capitalized on users’ willingness to pay to listen to expert sharing, making a splash in the knowledge service market. For example, a Live session by a famous doctor on ‘Cancer Prevention Course’ attracted tens of thousands of paid listeners, generating hundreds of thousands in revenue\u2014something unimaginable in the past. The result of the commercialization kickoff was that Zhihu achieved considerable revenue for the first time that year, breaking through from ‘zero revenue’ to ‘ten-million-level revenue,’ and also proved the commercial logic that users are willing to pay for knowledge. However, commercialization also brought conflicts with user experience: some old users were dissatisfied with the appearance of ads and soft articles on Zhihu, feeling the community had ‘changed its flavor.’ Zhou Yuan admitted in an interview that whenever someone asked how Zhihu makes money, he felt annoyed internally but had no choice but to explore ways to make money. He once joked, ‘Just make something commercial, and you make money,’ a quip that was later proven not easy\u2014commercialization is like a double-edged sword, hanging over Zhihu’s head. Despite the challenges, the 2016 commercialization turning point undoubtedly opened new growth space for Zhihu, moving it away from the ‘burn money to acquire users’ model and toward a ‘blood-generating cycle.’ In the long run, without the brave exploration of advertising and knowledge payment in 2016, Zhihu might have struggled to support its later IPO.<\/p>\n 3. 2018 Content Regulatory Storm: The Game Between Compliance and Freedom, Adjusting Community Operation Strategies. In March 2018, Zhihu was summoned by regulatory authorities due to the appearance of illegal and non-compliant information on the platform, leading to a 7-day suspension for rectification. This event had a profound impact on Zhihu, representing a passive turning point in its development. The background was that as Zhihu’s user base expanded, speech content became harder to control, with some sensitive political topics and inappropriate remarks attracting regulatory attention. Under strong regulatory policies, any content platform must strengthen audits, or face the risk of shutdown. Zhihu quickly changed during this crisis: conducting large-scale self-inspections to clean up non-compliant content, strictly implementing content audit standards, and closely monitoring sensitive areas (politics, ideology, etc.). During the rectification, Zhihu even suspended some functions, showing its determination. The turning point result was that Zhihu successfully passed the rectification and was restored to the app stores, avoiding a life-or-death crisis, but also lost some users who pursued freedom of speech. Some users found that political questions on Zhihu almost disappeared, and some hot current affairs questions were quickly deleted, which was completely different from the relatively relaxed discussion atmosphere in the past. Zhihu was thus mocked as ‘Bianhu’ (a pun suggesting it tailors content to official preferences), seen as too accommodating to the government. But from the perspective of business survival, this was a compromise that had to be made. The long-term impact was that Zhihu explored a path of ‘prudent operation’: in the Chinese policy environment, it preferred to lose some discussion freedom to preserve the overall community. Zhihu subsequently established a content risk control team, invested in technical forces to filter illegal content, and regularly reported rectification to the Cyberspace Administration of China. It can be said that the 2018 regulatory event marked Zhihu’s transition from ‘barbaric growth’ to ‘compliant operation.’ Under the new normal of strong regulation, Zhihu was able to survive safely and continue growing, which was pleasing to both users and investors. However, on the other hand, this move also changed the tone of the Zhihu community, with an increase in the proportion of entertainment and lifestyle content, while public topics and sharp discussions decreased. Some core users and top answerers were lost, and the overall content depth was criticized for declining. This change temporarily weakened Zhihu’s differentiated advantage. However, over time, Zhihu has gradually regained some space for serious discussions through professional certifications and institutional accounts (such as the ‘Blue V Professional Identity Certification’ launched in 2023). In summary, the 2018 content compliance turning point was a painful period in Zhihu’s development, but after rising from the ashes, Zhihu’s risk resistance and standardization levels improved, and it also cleared policy obstacles for its later listing.<\/p>\n 4. 2021 U.S. IPO: Embracing the Capital Market, Entering a New Era of Development. After ten years of growth, Zhihu finally embarked on its IPO journey in 2021. The background for choosing to list on the U.S. stock market was that Chinese concept stocks generally had decent valuations at the time, and Zhihu had been losing money for many years, needing substantial funds to support long-term operations and new business expansion. The reasons for the IPO were clear: to raise development funds for the company, provide an exit channel for early investors, and enhance brand image and credibility. Zhihu successfully listed on the New York Stock Exchange in March, becoming the ‘first Chinese Q&A stock.’ On the first day of listing, its market capitalization was about $5.3 billion, with shareholders like Tencent and Innovation Works seeing considerable paper gains. The direct result of the listing was that Zhihu raised over $500 million in one go, greatly alleviating financial pressure. Zhihu primarily used these funds to increase technology investment, build content ecosystems, and incubate new businesses (such as vocational education). In the subsequent 2021\u20132022 period, relying on the funds raised from the listing, Zhihu expanded its team and developed underlying technologies such as algorithm recommendations, live streaming, and AI recognition, strengthening community infrastructure. Additionally, the IPO boosted Zhihu’s brand awareness, bringing more business collaboration opportunities, as many large clients were more willing to engage in advertising and content co-creation with the listed company Zhihu. It can be said that the listing truly propelled Zhihu into a new stage of ‘capital + industry’ dual-wheel driven development. In terms of long-term impact, the listing was both an opportunity and a challenge. The opportunity was that Zhihu stepped onto a larger stage, with expanded financing channels (subsequent bond issuances, additional share offerings, etc.), while corporate governance gradually aligned with international standards, becoming more transparent and standardized. However, the challenge was that as a listed company, Zhihu had to disclose quarterly results, and from then on, it was exposed to the spotlight, with unprecedented pressure on growth and profitability. Especially amid the downturn in Chinese concept stocks and the expansion of its own losses (Zhihu’s stock price once fell below its IPO price after listing, raising doubts), management had to prove the sustainability of its business model to shareholders. This pressure actually catalyzed Zhihu’s strategic shift towards cost reduction, efficiency improvement, and focusing on core businesses after 2022. In summary, the 2021 listing was a milestone turning point in Zhihu’s development history, marking its transformation from being driven by VC investment to operating under public shareholder supervision. Zhou Yuan remarked at the time, ‘After eleven years, Zhihu finally has its moment in the spotlight, but business challenges remain to be solved.’ It is foreseeable that the standardized governance and brand effects brought by the listing will benefit Zhihu in the long term, while the high demands from capital will force Zhihu to continuously innovate and evolve.<\/p>\n 5. 2022\u20132023 Strategic Contraction and Transformation: Focusing on High-Value Businesses, Embarking on the Path to Profitability. After the rapid expansion following the listing and the massive losses in 2021, Zhihu began a proactive strategic turning point in 2022. The characteristic of this turning point was not a major offensive into new areas, but rather tightening the front lines, improving quality and efficiency. The background was the macroeconomic downturn and the company’s low stock price, requiring Zhihu to change its ‘loss-for-growth’ model. Thus, Zhihu’s management decisively adopted several measures: cutting or weakening businesses with little synergy with the core Q&A business (such as short video entertainment content), and concentrating resources on developing two high-margin businesses: paid memberships and vocational education. This meant Zhihu gave up competing with Douyin in short content, instead strengthening its moat in knowledge payment. For example, it laid off most of the video team and tilted content creation incentive budgets more towards high-quality text and course creators. Meanwhile, it made large investments in the vocational education sector, rapidly supplementing the course product line through investments and acquisitions, aiming to create a profitable education sub-brand within 2\u20133 years. Zhihu also initiated an internal ‘Deep Blue’ project to conduct efficiency reviews of operations, marketing, and human resources departments, reducing redundancy and internal friction. A series of contraction actions were intensively implemented from the second half of 2022 to early 2023, with obvious short-term effects: Zhihu’s operating expenses in Q3 2023 decreased by 30% year-on-year, and net losses nearly broke even. It can be said that Zhihu finally found the feeling of ‘doing more with less,’ no longer burning money to buy growth as in the past. The result of this turning point was that Zhihu successfully achieved its first quarterly profit by the end of 2023, proving the correctness of the cost reduction and efficiency improvement strategy. More importantly, the strategic consensus within the company became clearer\u2014Zhihu will long-term focus on its core community and knowledge services, no longer blindly diversifying. Previously high-hoped but ineffective sideline businesses can be decisively stopped, and resources should be used in areas where it is truly skilled and promising. In terms of long-term impact, this transformation made Zhihu’s organization leaner and more efficient, its business structure healthier, and also restored investor confidence in Zhihu (as evidenced by the stock price stabilizing and rising in the fourth quarter of 2023). However, strategic contraction is not without cost: Zhihu’s monthly active user (MAU) growth nearly stagnated, hovering around 100 million for two years, with no significant increase in market share. In terms of competitive landscape, Zhihu’s proactive withdrawal from the short video user competition means that acquiring new users from Generation Z in the future may rely more on other methods (such as attracting through professional courses). Overall, the strategic adjustment in 2022\u20132023 was a wise choice for Zhihu to ‘endure short-term pain for long-term gain,’ opening a realistic path to profitability for the company. This turning point also meant that Zhihu entered a stage of mature operations, bidding farewell to extensive growth and stepping into a new normal of intensive cultivation.<\/p>\n 6. 2024 Turning Losses into Profits: Validating the Business Model, Achieving Key Breakthrough. After the series of adjustments mentioned above, Zhihu finally achieved its first overall profit in the fourth quarter of 2024. This marks the answer to Zhihu’s 13-year ‘profitability challenge,’ and can be considered a milestone event in its strategic turning points. The important background for profitability is Zhihu’s years-accumulated content ecosystem and user loyalty, which, after continuous optimization of monetization methods, finally released profits. The direct reasons are: on one hand, a healthier revenue structure\u2014in 2024, membership subscriptions and vocational education revenue already accounted for more than half, improving the overall gross margin; on the other hand, costs and expenses were strictly controlled, with sales and marketing expenses dropping to a historical low proportion. The result is that despite a year-on-year decline in revenue, Zhihu was still able to achieve profitability, demonstrating the strong resilience of its business model. The first profit is of great significance for Zhihu’s medium- to long-term development: first, it proves that Zhihu can turn losses into profits like Bilibili, breaking the external doubt that ‘Zhihu can’t make money.’ Once profitability is verified, the capital market’s perception of Zhihu will change positively, and the company’s valuation is expected to recover. Second, with profits, Zhihu has more strategic options. It can choose to invest more funds in long-term projects (such as AI R&D) without worrying about excessive cash burn; it can also consider rewarding shareholders, promoting positive stock price performance, and thus binding and attracting top talents through equity incentives. Profitability also means a shift in Zhihu’s internal assessment orientation\u2014from purely pursuing KPI growth to a more balanced consideration of input-output ratios, which helps build a more rational corporate culture. Finally, profitability positions Zhihu favorably in the fierce competition for existing internet users. Many community products are forced to shut down or sell due to inability to profit, but Zhihu has survived to profitability, giving it more confidence for independent development. Of course, the first profit is just the beginning; how to stabilize and expand profitability is the next topic. Zhihu executives are soberly aware that industry-leading net profit margins have not yet been achieved, and there is still much room for improvement. Therefore, the long-term impact of the profitability inflection point depends on whether Zhihu can continuously innovate and expand profit sources, rather than just profiting through cost-saving. In summary, turning losses into profits in 2024 is a glorious turning point in Zhihu’s development history, signifying the maturity of its business model and heralding a new stage of profitable growth in the future. At this moment, Zhou Yuan can finally breathe a sigh of relief\u2014the ’14-year-old veteran community has finally crossed the breakeven point.’<\/p>\n (Among the strategic turning points listed above, points 1, 2, 4, 5, and 6 are changes caused by Zhihu’s active decisions, while point 3 (content regulation) is a passive change driven by external factors. However, regardless of the type, these events collectively shape Zhihu’s current form. Zhihu has moved from niche to mass, from idealism to reality, from expansion to contraction, and from losses to profitability. Each turning point has been accompanied by pain and choices, but also laid the foundation for long-term development. Looking ahead, Zhihu will still face new turning points and challenges, such as how to deal with the impact of AI on the Q&A model, which requires management to assess the situation and continue the decisiveness and execution demonstrated at key nodes in the past.)<\/p>\n #### Changes in Core Competitiveness and Competitor Comparison<\/p>\n Since its establishment in 2010, Zhihu’s core competence has continuously evolved and adjusted with changes in internal and external environments. At different stages, Zhihu’s core capability focus and differentiation paths from main competitors are as follows:<\/p>\n #### References:<\/p>\n #### Basic Information<\/p>\n Zhihu (formerly known as"Zhihu"), is China’s leading online Q&A community and knowledge sharing platform, officially founded by Zhou Yuan on January 26, 2011. Zhou Yuan, born on December 18, 1980, in Guiyang, Guizhou, graduated from Chengdu University of Technology with a bachelor’s degree and from Southeast University with a master’s degree in Software Engineering [10]. Before founding Zhihu, Zhou Yuan worked as a programmer in Shanghai and later as a journalist in Beijing, with extensive experience in the internet and media industries [11].<\/p>\n Zhihu’s initial positioning was to provide users with a high-quality knowledge sharing platform, with its slogan being"Have a question? Ask Zhihu"The platform adopted an invitation-based registration system, with a small initial user base, primarily concentrated in the internet industry [45].<\/p>\n ###### Company IPO Timing, Valuation, and Listing Location<\/p>\n Zhihu went public on the New York Stock Exchange on March 26, 2021, with the stock code"ZH". The IPO price was set at $9.5 per ADS (American Depositary Share), raising a total of $850 million (after the full exercise of the over-allotment option) [74]. Prior to the IPO, Zhihu had completed 8 rounds of financing, raising approximately $920 million in total [25].<\/p>\n #### Pre-IPO Development (2011-2021)<\/p>\n 2011: Closed Beta Period<\/p>\n Zhihu officially launched on January 26, 2011, adopting an invitation-based registration system. Early users included prominent internet figures such as Kaifu Lei, Huateng Ma, Xing Wang, Xiaochuan Wang, and Xiaoping Xu [30]. During this period, Zhihu positioned itself as an"internet"-driven platform, with a user base in the hundreds of thousands [45].<\/p>\n Zhihu’s initial model borrowed from the American Q&A platform Quora, but incorporated Chinese characteristics. The platform focused on high-quality Q&A content, encouraging users to share knowledge and insights, forming a unique community culture [17].<\/p>\n 2013-2014: Early Expansion After Opening Up<\/p>\n In 2013, Zhihu began open registration, and the user base rapidly grew from 400,000 to 4 million [45]. On October 28, 2014, Zhihu welcomed its 10 millionth registered user, marking the platform’s entry into a phase of rapid development [45].<\/p>\n During this period, Zhihu’s business focus was on expanding the user base and optimizing content quality, gradually extending from the closed internet circle to broader fields. Zhihu began experimenting with commercialization, launching an advertising-based revenue model in 2014 [26].<\/p>\n 2015-2017: Explosive Growth During the Open Period<\/p>\n In 2015, Zhihu received a $55 million Series C financing round led by Tencent, with user numbers exceeding 50 million [45]. That same year, Zhihu began expanding into paid knowledge services, launching"Zhihu Value" (paid Q&A),"Zhihu Live", and other products [40].<\/p>\n In September 2017, Zhihu’s registered users surpassed 100 million, and it received a $100 million Series D financing round led by Today Capital [45]. During this period, Zhihu’s commercialization strategy became more diversified, with paid knowledge services becoming a significant revenue source.<\/p>\n 2018-2019: Intelligent Community Phase<\/p>\n In 2018, Zhihu’s commercialization became further diversified. In April, Zhihu launched its e-book product"Zhihu Reading Club". In June, it launched a bundled paid membership service"Zhihu University" [47]. In 2018, Zhihu’s registered users reached 160 million, with the number of answers exceeding 100 million [45].<\/p>\n In August 2019, Zhihu completed its Series F financing round, raising a total of $434 million, led by Kuaishou and followed by Baidu, with existing investors Tencent and Today Capital continuing to invest [79]. This was Zhihu’s largest financing round to date and one of the largest financings in the Chinese internet culture and entertainment sector [91].<\/p>\n ###### Financing History and Valuation<\/p>\n From January 2011 to August 2019, Zhihu conducted 8 rounds of financing from the angel round to the Series F round, raising a total of approximately $920 million [25]. Below are Zhihu’s major financing events:<\/p>\n Angel Round: 2011, received angel investment from Innovation Works<\/p>\n Series A: 2012, received nearly $10 million investment from Qiming Venture Partners<\/p>\n Series B: June 2014, received $22 million investment from SAIF Partners and Qiming Venture Partners<\/p>\n Series C: September 2015, received $55 million investment from Tencent and Sogou, with a valuation of approximately 5 billion RMB<\/p>\n Series D: January 2017, received $100 million investment from Today Capital<\/p>\n Series E: July 2018, received $300 million investment, with a valuation approaching $2.5 billion<\/p>\n Series F: August 2019, received $434 million investment<\/p>\n ###### Changes in User Scale<\/p>\n Zhihu’s user scale has experienced rapid growth, from hundreds of thousands of users in 2011 to 10 million users in 2014, then to 100 million users in 2017, and finally to 160 million users in 2019 [45]. As of June 2019, Zhihu’s user base reached 280 million [45].<\/p>\n Zhihu’s users are mainly knowledge-based middle-class individuals, with a large proportion of young people, primarily located in developed coastal regions, and over 80% of users have a bachelor’s degree or above [45]. These high-quality users provide a solid foundation for Zhihu’s content quality and commercialization.<\/p>\n ###### Core Product Iterations<\/p>\n Zhihu’s core products have undergone multiple iterations and feature launches:<\/p>\n Early Products: When launched in 2011, Zhihu used a pure text format; image functionality was not added until spring 2012 [53].<\/p>\n Paid Knowledge Products: In 2016, launched"Zhihu Value" (paid Q&A),"Zhihu Live"\u3001"Zhihu Bookstore" and other products [40].<\/p>\n Paid Membership Services: In 2018, launched"Zhihu University" bundled paid membership service [47].<\/p>\n E-book Products: In April 2018, launched"Zhihu Reading Club" e-book products [47].<\/p>\n AI Technology Applications: In 2018, began applying AI technology in content recommendations and community management.<\/p>\n ###### Evolution of Commercialization Strategies<\/p>\n Zhihu’s commercialization strategies have gone through three main phases:<\/p>\n Advertising Model: Launched an advertising-based revenue model in 2014 [26].<\/p>\n Paid Knowledge: From 2016 onwards, gradually expanded paid knowledge services, including"Zhihu Value"\u3001"Zhihu Live" and other products [40].<\/p>\n Diversified Commercialization: In 2018, commercialization became further diversified, launching"Zhihu Reading Club"\u3001"Zhihu University" and other products [47].<\/p>\n #### Post-IPO Development (2021 to Present)<\/p>\n ###### Financing Performance and Market Capitalization Changes<\/p>\n Zhihu went public on the New York Stock Exchange on March 26, 2021, with an IPO price of $9.5 per ADS, raising a total of $850 million. On its first day of trading, Zhihu’s stock performed poorly, dropping over 20% after opening and closing down 11.16% at $8.44, with a total market capitalization of $4.718 billion [3].<\/p>\n Excluding the green shoe over-allotment, Zhihu’s IPO raised approximately $520 million, with a valuation of $5.3 billion, an increase of over 50% from its last financing valuation [8]. After listing, Zhihu’s market capitalization has fluctuated but has generally remained at a certain level.<\/p>\n ###### Financial Situation Analysis<\/p>\n Based on Zhihu’s 2022 annual report data:<\/p>\n Revenue: Zhihu’s revenue primarily comes from advertising, paid memberships, commercial content solutions, and other services.<\/p>\n Losses: Zhihu is still operating at a loss, but the loss rate has narrowed.<\/p>\n Cash Flow: Zhihu’s operating cash flow has improved, indicating better financial health.<\/p>\n In 2022, Zhihu’s revenue structure underwent significant changes, with commercial content solutions accounting for a larger share of revenue and becoming an important revenue source. Meanwhile, paid membership revenue also grew steadily.<\/p>\n ###### Business Focus Adjustments<\/p>\n After listing, Zhihu’s business focus has undergone multiple adjustments:<\/p>\n Content Ecosystem Development: Zhihu increased its investment in the content ecosystem, launching initiatives such as the"Haiyan Plan" to improve content quality.<\/p>\n AI Technology Applications: In 2023, Zhihu increased its investment in AI technology, launching AI-related products and services.<\/p>\n Internationalization Strategy: Zhihu began exploring international markets, attempting to replicate its successful model globally.<\/p>\n Commercialization Upgrade: Zhihu further optimized its commercialization strategies, improving the quality and effectiveness of commercial content solutions.<\/p>\n ###### Major Personnel Changes<\/p>\n After listing, Zhihu experienced several important personnel changes:<\/p>\n Management Team Adjustments: Some executive positions were adjusted to adapt to the new stage of company development.<\/p>\n Layoffs: In 2023, Zhihu conducted a small-scale layoff, primarily affecting certain business departments.<\/p>\n ###### Transformation in the AI Era<\/p>\n In 2023, with the rapid development of AI technology, Zhihu also began its layout in the AI field:<\/p>\n AI Product Launch: Zhihu launched new products and services based on AI technology.<\/p>\n Content Quality Improvement: Utilized AI technology to enhance content quality and reduce low-quality content.<\/p>\n User Experience Optimization: Optimized user recommendations and content display through AI technology to improve user experience.<\/p>\n #### External Investment and M&A Activities<\/p>\n ###### Major External Investment Projects<\/p>\n Before its IPO, Zhihu had already conducted multiple financing rounds, but its external investment and M&A activities were relatively limited. Main investment activities focused on the content ecosystem and related fields.<\/p>\n ###### Evolution of Investment Strategy and Logic<\/p>\n Zhihu’s investment strategy primarily revolves around the following aspects:<\/p>\n Content Ecosystem Development: Investing in high-quality content creators and content platforms.<\/p>\n Technology Upgrade: Investing in areas such as AI technology and data analysis.<\/p>\n International Expansion: Attempting to expand into international markets through investment and cooperation.<\/p>\n ###### Assessing the Impact of Investment Activities<\/p>\n Zhihu’s investment activities have had a positive impact on the company’s overall business:<\/p>\n Content Quality Improvement: By investing in high-quality content creators, the platform’s content quality has been enhanced.<\/p>\n Technology Capability Enhancement: Through investments in AI and technology fields, the company’s technical capabilities have been strengthened.<\/p>\n International Layout: Laid the foundation for future international expansion.<\/p>\n #### Analysis of Key Turning Points<\/p>\n There are several important strategic turning points in Zhihu’s development process:<\/p>\n From Closed to Open: In 2013, Zhihu transitioned from invitation-based registration to open registration, leading to a rapid expansion of its user base.<\/p>\n Exploration of Paid Knowledge: In 2016, Zhihu began exploring the paid knowledge model, launching"Zhihu Value"\u3001"Zhihu Live" and other products.<\/p>\n Diversified Commercialization: In 2018, Zhihu launched"Zhihu University" and other diversified commercial products, enriching revenue sources.<\/p>\n IPO Financing: In 2021, Zhihu went public on the NYSE, securing substantial funding to lay the foundation for future development.<\/p>\n AI Technology Layout: In 2023, Zhihu began its layout in the AI technology field, adapting to industry development trends.<\/p>\n #### Changes in Core Competitiveness<\/p>\n ###### Initial Competitiveness<\/p>\n In the early stage, Zhihu’s core competitiveness was mainly reflected in:<\/p>\n High-Quality Content: Through invitation-only registration and high-quality users, content quality was ensured.<\/p>\n Community Culture: A unique community culture and discussion atmosphere were formed.<\/p>\n Prominent Users: Attracted a large number of well-known figures from the internet industry, enhancing the platform’s influence.<\/p>\n ###### Post-IPO Competitiveness<\/p>\n After listing, Zhihu’s core competitiveness changed:<\/p>\n Technology-Driven: Greater emphasis on being technology-driven, particularly in the layout of the AI field.<\/p>\n Content Ecosystem: A more comprehensive content ecosystem has been built.<\/p>\n Commercialization Capability: Enhanced commercialization capabilities and diversified revenue streams.<\/p>\n User Scale: The user base has further expanded, covering a broader range of groups.<\/p>\n #### Differentiation from Major Competitors<\/p>\n Comparison of Zhihu’s differentiation strategies with major competitors:<\/p>\n #### Conclusion<\/p>\n Since its founding in 2011, Zhihu has gone through 14 years of development. From the initial closed Q&A community to a pre-IPO knowledge-sharing platform, and then to a post-IPO diversified content community, Zhihu has undergone multiple strategic transformations and business adjustments. By continuously optimizing content quality, enhancing user experience, and expanding commercialization channels, Zhihu has successfully evolved from a small internet community into a large content platform with hundreds of millions of users.<\/p>\n Against the backdrop of rapid AI technology development, Zhihu is actively positioning itself in the AI field, laying the foundation for future development. Despite facing some challenges after its IPO, Zhihu has maintained strong market competitiveness and growth potential. In the future, Zhihu will continue to optimize its content ecosystem, enhance commercialization capabilities, and provide users with a higher-quality knowledge-sharing platform.<\/p>\n #### References<\/p>\n Based on the two research reports, there are multiple contradictions and conflicts in the content, mainly focusing on the following aspects:1. Contradictions in Basic Information<\/strong> Founding Time: Report A states that Zhihu was"officially founded on January 26, 2011", while Report B states that"Zhihu was founded in 2010", established in Beijing in August 2010, entered internal testing in December, and was"officially launched on January 26, 2011" Founding Team: Report A only mentions Zhou Yuan as the sole founder, while Report B states that"Zhou Yuan and partners like Zhang Liang founded Zhihu in Beijing", with the founding team including"CEO Zhou Yuan, CTO Li Shenshen, COO Huang Jixin, etc." 2. Contradictions in IPO Financing and Listing Information<\/strong> IPO Funds Raised: Report A states"a total of $850 million was raised (after the full exercise of the over-allotment option)", while Report B states"approximately $522 million was raised" Hong Kong Listing: Report A completely omits the important event of Zhihu’s secondary listing in Hong Kong, while Report B details that"Zhihu achieved a dual primary listing on the Main Board of the Hong Kong Stock Exchange on April 22, 2022 (Stock Code: 2390)" 3. Contradictions in Financial Situation<\/strong> Current Profit\/Loss Status: Report A states"Zhihu is still operating at a loss, but the loss rate has narrowed", while Report B clearly indicates that"Zhihu achieved its first full profitability in the fourth quarter of 2024: revenue of 860 million yuan for the quarter, net profit of 86.4 million yuan" Detail Level of Financial Data: Report A provides extremely limited financial data, only generally mentioning revenue sources; Report B provides complete revenue and loss data for 2020-2024.4. Contradictions in Strategic Focus and Business Layout<\/strong> Core Business Development: Report A emphasizes"in 2023, Zhihu increased its investment in the AI technology field", while Report B highlights"Zhihu has identified'vocational education'as its second growth curve" Business Adjustments: Report A briefly mentions"Zhihu’s business focus has undergone multiple adjustments", while Report B details that in 2022-2023, Zhihu actively"streamlined its operations, improved quality and efficiency"\uff0c"by cutting or downplaying businesses that did not synergize well with its core Q&A business." 5. Contradictions in External Investment and M&A<\/strong> Investment Activities: Report A states"Zhihu’s external investment and M&A activities are relatively few", without listing specific cases; Report B, however, details Zhihu’s investments and acquisitions in vocational education companies such as Pinzhi Education, Papa Education, and Palm Gardener.6. Contradictions in Key Strategic Turning Points<\/strong> Content Regulation Incident: Report A completely omits the content regulation storm in 2018; Report B considers it an important turning point:"In March 2018, Zhihu was interviewed by regulatory authorities due to the presence of illegal information on the platform and was taken down for rectification for 7 days." Profitability Turning Point: Report A does not mention the major breakthrough of Zhihu achieving profitability; Report B lists"achieving profitability in 2024"as a key strategic turning point.7. Contradictions in Post-IPO Development Trajectory<\/strong> Business Adjustments: Report A’s description of Zhihu’s post-IPO business adjustments is vague; Report B details that in 2022, Zhihu"significantly cut its video team" and implemented strategic contraction measures such as reducing marketing expenses."Report Timeliness: Report B covers situations up to"the first quarter of 2025", while Report A only vaguely mentions"to dateand does not include the key changes of 2024.<\/strong> Competitor Comparison: Report A only briefly mentions Xiaohongshu, Bilibili, and WeChat Official Accounts; Report B provides a detailed analysis of differentiated competitive strategies with Baidu, Weibo, Xiaohongshu, and Bilibili. Competitiveness Evolution: Report A’s competitiveness analysis is relatively static; Report B provides a detailed analysis of Zhihu’s evolution from"a high-quality Q&A community atmosphere"to"full-chain knowledge content services". These contradictions indicate that Report B provides more detailed and updated information in most aspects, especially regarding post-IPO development, strategic transformation, and financial situation, while Report A has omissions or cursory treatment of many important events.<\/p>\n First, the two reports are consistent in the data they researched,and no errors occurred.<\/strong>\u3002<\/p>\n Here, I need to explain Claude’s Point 2:The inconsistency in the IPO funds raised.<\/strong>\u3002<\/p>\n AutoGLM mentions a total of $850 million raised, sourced from Yicai, with the original text noting in parentheses after $850 million (after the full exercise of the over-allotment option). Simultaneously, AutoGLM also mentions ‘Excluding the green shoe over-allotment, Zhihu’s IPO financing was approximately $520 million,’ which aligns with the $522 million figure researched by OpenAI.<\/p>\n As for whether the over-allotment was exercised at the time, I was completely unaware, so I sought help from the onlookers Grok and DeepSeek.<\/p>\n Both external aids conducted in-depth research and reached a consistent conclusion: the underwriters did exercise the over-allotment option when Zhihu was listed on the NYSE. This indicates that OpenAI’s DeepResearch failed to research the over-allotment situation at the time. AutoGLM gets a point.<\/p>\n However, this is AutoGLM’s only point gain.<\/p>\n Similarly, in the IPO section, a very crucial piece of information,Zhihu’s secondary listing in Hong Kong, was completely overlooked by AutoGLM.<\/strong>\u3002<\/p>\n And, aside from this, for the other 7 points, AutoGLM’sresearch results are less detailed than OpenAI’s.<\/strong>Especially for thefinancial report data explicitly required in the Prompt, as public data, AutoGLM failed to research it, which is a significant point of loss.<\/strong>. Also, AutoGLM’s research results lack timeliness; despite conducting online searches, it only researched up to 2023. In its research report, Zhihu is still in a continuous loss-making state.<\/p>\n In terms ofcomprehensive strength for in-depth research, there is still a clear gap compared to OpenAI.<\/strong> It is worth noting that while OpenAI’s sources are only 33, most come from financial media and industry reports.<\/p>\n As for AutoGLM, it reviewed over 90 sources, buta large portion of them were UGC content<\/strong>. The highest proportion of information sources isZhihu<\/strong>, with much of it being reposted information from minor accounts.<\/p>\n This is not to say that Zhihu’s UGC content is bad, but first, the ratio of official media to UGC content is worth optimizing; second, this type of financial news press release is not typically the domain where Zhihu’s professional respondents are active. Most search results would be from spam accounts and marketing accounts reposting, and it does not truly utilize the high-quality content on the Zhihu platform. This is worth AutoGLM reconsidering.<\/p>\n Since AutoGLM operates by controlling a browser, some social media sites (including Zhihu) that cannot be accessed via API or require login expose their search operation process to the user.<\/p>\n Observing its operations, one point worth acknowledging is that it is able touse functions like sorting by time, sorting by most liked, and viewing only articles to improve accuracy<\/strong>. However, in terms ofsearch habits, there is definitely room for improvement<\/strong>. AutoGLM’s search habits are those of very early search engines. Rigidly applying this search method to content communities can onlyyield a large number of irrelevant answers<\/strong>\u3002<\/p>\n . However, AutoGLM’s information filtering is acceptable, as these irrelevant pieces of information did not appear in the final report. But the search habits are still worth optimizing; first, to save tokens, and second, optimized individual research results might be more in-depth.<\/p>\n Although AutoGLM still has a gap compared to SOTA, its advantage lies in being free and capable of automatic browser operations. Smart people have likely already found its uses. Moreover, the current AutoGLM Deep Research version has shown significant improvement compared to the previous ‘NiuNiu’. It is believed that with continued iteration and optimization, it will become even more user-friendly.<\/p>\n","protected":false},"excerpt":{"rendered":" A practical evaluation of Zhipu’s AutoGLM DeepResearch tool, tested by analyzing Zhihu’s 14-year business history and compared with OpenAI’s DeepResearch.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[24],"tags":[15,23,14],"class_list":["post-138","post","type-post","status-publish","format-standard","hentry","category-articles","tag-ai-agent","tag-deepseek","tag-14"],"_links":{"self":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts\/138","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/comments?post=138"}],"version-history":[{"count":2,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts\/138\/revisions"}],"predecessor-version":[{"id":1032,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/posts\/138\/revisions\/1032"}],"wp:attachment":[{"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/media?parent=138"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/categories?post=138"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/en.blog.liu-qi.cn\/index.php\/wp-json\/wp\/v2\/tags?post=138"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Research Task: Commercial Analysis of ‘Zhihu’<\/h2>\n
Prompt<\/h3>\n
Please conduct a comprehensive research analysis on the development history of Zhihu Company from its establishment to the present (approximately 14 years) from the perspective of a professional business analyst. Divide the analysis into two main stages: pre-IPO and post-IPO. Please provide detailed information on the following: ## Basic Information- Zhihu's founding date, founders, and initial positioning- The company's IPO time, valuation, and listing location ## Pre-IPO Development (by year)- Main business focus and model changes each year- Key product iterations and feature launches- Major financing events and investors- User scale changes and key growth milestones- Evolution of commercialization strategies ## Post-IPO Development (by year or quarter)- Quarterly\/annual financial performance (revenue, profit\/loss, cash flow)- Main revenue sources and changes in their proportions- Business focus adjustments and strategic transformations- Market capitalization changes and investor reactions- Major personnel changes (including layoff events) ## External Investment and M&A Activities- List Zhihu's major external investment projects and amounts over the years- Analyze the evolution of Zhihu's investment strategy and logic- Distribution of investment fields and synergy with core business- Analysis of major acquisition cases and integration effects- Evaluate the impact of investment activities on Zhihu's overall business ## Analysis of Key Turning Points- List 5-7 of the most important strategic turning points in Zhihu's history- Analyze the background, decision reasons, and results of each turning point- Evaluate the impact of each turning point on the company's long-term development ## Changes in Core Competitiveness- Analyze how Zhihu's core competitiveness has evolved over time- Comparative analysis of differentiation strategies with major competitors. Please ensure the analysis is objective and data is accurate, citing reliable sources. If possible, add key financial indicator charts to enhance the visual intuitiveness of the analysis. The final analysis should present the complete commercial development trajectory of Zhihu over 14 years, with special attention to the evolution of its business model and changes in profitability.<\/code><\/pre>\nOpenAI’s Research Results<\/h2>\n
\n
![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/001-433f211702ee.png\")
<\/p>\n\n
![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/002-487ef6d6a8d0.png\")
<\/p>\n\n
![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/003-f988b3b3b4aa.png\")
A total of 33 sources:<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/004-cf176c52f0da.png\")
<\/p>\nResearch results from AutoGLM Pondering Edition<\/h2>\n
![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/005-5733ec5bdad6.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/006-6a17fb6e1287.png\")
<\/p>\nJudge Claude conducts evaluation.<\/h2>\n
Prompt<\/h3>\n
I have two research reports on Zhihu here. Please help me compare their research conclusions (including details) to identify any contradictions or conflicts. This is Report A: AutoGLM Deep Research Version. This is Report B: OPENAI Version, accompanied by charts generated by B.<\/code><\/pre>\nJudge’s Remarks<\/h3>\n
Summary Time<\/h2>\n
Grok’s Reply<\/h3>\n
![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/007-40f151268ca9.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/008-eb069e4e40f0.png\")
<\/p>\nDeepSeek’s Reply<\/h3>\n
![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/009-284b3fdca01c.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/010-cd04064e8a69.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/011-ba8e94786124.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/012-bb85ff8e0312.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/013-47d596a06224.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/014-8a63e5d38f1b.png\")
<\/p>\n![\"\"](\"https:\/\/blog.liu-qi.cn\/wp-content\/uploads\/2025\/04\/015-c36e7324339d.png\")
<\/p>\n