Q1 2025 Kingsoft Cloud Holdings Ltd Earnings Call

Unknown Executive: Good day, and thank you for standing by.

Unknown Executive: Welcome to the Kingsoft Cloud's first quarter 2025 earnings conference call. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question and answer session. To ask a question during the session, you will need to press star, 1, 1 on your telephone. You will then hear an automatic message advising your hand is raised. To withdraw your question, please press star, 1, 1 again. Please be advised that today's conference is being recorded.

Nicole Shan: I would now like to hand the conference over to your speaker today, Nicole Shan. I am the Director of Kingsoft Cloud. Please go ahead.

Unknown Executive: Thank you, operator.

Nicole Shan: Hello, everyone, and thank you for joining us today. Kingsoft Cloud's first quarter 2025 earnings release was distributed earlier today and is available on our IR website at ir.ksyun.com, as well as on the PR Newsletter services. They will be available to answer your questions during the Q&A session that follows. There will be consecutive interpretations. Our interpretations are for your convenience and the reference purpose only. In case of any discrepancy, management statements in the original language will prevail.

Unknown Executive: Before we begin, I'd like to remind you that this conference call contains forward-looking statements within the meaning of Section 21E of the Security Exchange Act of 1934, as demanded and as defined in the U.S. Private Security Litigation Reform Act of 1995. These forward-looking statements are based upon management's current expectation and current market and operating conditions and relate to events that involve known or unknown risks, uncertainties, and other factors, all of which are difficult to predict and many of which are beyond the company's control, which may cause the company's actual results, performance, or achievements to differ maturely from those in the forward-looking statements.

Unknown Executive: Further information regarding this and other risk uncertainty or factors are included in the company's findings with the U.S. as the company does not undertake any obligation to update any forward-looking statement as a result of new information, future events, or otherwise, except as required under applicable law.

Unknown Executive: Finally, please note that unless otherwise stated, all financial figures mentioned during this conference call are denominated in RMB.

Nicole Shan: It's now my pleasure to introduce our Vice Chairman and CEO, Mr. Zhao. Please go ahead. Thank you.

Tao Zou: Welcome to Kingsoft Cloud's 1st Group Call of the Year 2025 I am CEO of Kingsoft Cloud, Tao Zou This quarter, we will continue to strongly promote our company's business. 以高质量可持续发展为目标 聚焦AI关键领域 So, yeah Our income has increased by 11%. $19.7 billion Gongyou Cloud and Hangye Cloud have achieved the same growth. Among them, the number of employees increased by 14%. with a net income of 13.5 billion yuan. Next, we will continue to improve with AI guidance.

Tao Zou: And thank you for tuning into the civics talk broadcast. The business bill is 5.3 billion yuan. 同比增长超过200% Huanbi increased by 11% Zhan Gongyouyun, income proportion, continue to improve 39%. In addition.. This quarter, we are using a more flexible way of investing funds to accelerate the construction of computing machines. will be available in the second quarter. to further accelerate the improvement of AI income.

Tao Zou: Third, as Xiaomi Jinshan Ecology's only strategic cloud platform. We are moving forward with ecological business cooperation. This quarter, Xiaomi and Jinshan Ecology received a revenue of 500 million yuan. have increased by 50%. Our net income has increased by 25%. Our cooperation with Xiaomi Jinshan in the field of AI continues to advance and is getting deeper. to fully integrate the advantages and resources of the two sides and jointly expand the cloud infrastructure of the AI era with Xiaomi Jinshan. Finally, from a profitable point of view, After the adjustment of this quarter, Maoli is 3.3 billion yuan.

Tao Zou: The growth rate is 9.6% After adjustment, EBITDA reached 16.2% with an increase of 14.3%, mainly due to the continuous increase in AI revenue. But we also realize that there is a fluctuation in profit-return ratio. Jing Tao Zheng Mao Li Li Huan Bi Xia Jia 2.6% to 16.6% The decline in interest rates is mainly due to the decline in the share price of the industry cloud income, which has led to a decrease in contribution profits. and the variable impact of computing costs. After the adjustment, the operating profit was affected by the decline in gross profit and lost 55.81 million yuan this quarter.

Tao Zou: After adjustment, the operating profit margin is 2.8%. 相较去年同期的亏损 Compared to the loss of the same period last year 7.2% increased by 4.4%, and the profit turned into a loss.

Tao Zou: 尽管本季度的财务业体短期内出现环比波动 market changes and supply chain factors But from a long-term perspective, we are firmly following the long-term good strategic layout. Wenbu Maijin In cooperation with Xiaomi Jinshan Ecology, we have successfully laid the foundation for AI application deployment. to promote the company's overall AI cloud service strategic layout.

Unknown Executive: Hello, everyone.

Tao Zou: Thank you and welcome all for joining Kingsoft Cloud's first quarter 2025 earnings call. I am Tao Zou, CEO of Kingsoft Cloud. This quarter, we continued to steadily advance our business to the target on high quality and sustainable development, centering on key areas of AI. First, we recorded year-over-year revenue growth of 11%, reaching RMB1.97 billion. Both public cloud and enterprise cloud achieved year-over-year growth, among which public cloud increased by 14%, reaching RMB 1.35 billion. Second, we continued to drive progress with AI. This quarter, AI gross billing reached RMB525 million, representing a year-over-year increase of over 200%, and a quarter-over-quarter growth of 11%.

Tao Zou: further contributing 39% of public cloud revenue. In addition, this quarter, we are accelerating the construction of our computing clusters with more flexible capital deployment, which is expected to launch service officially in the second quarter, further boosting our AI business revenue. Third, as the only strategic cloud platform of the Xiaomi and Kingsoft ecosystem, our business cooperation with the ecosystem progressed smoothly. This quarter, revenue from Xiaomi and Kingsoft ecosystem reached RMB 500 million, up 50% year over year, with its contribution to total revenue further increasing to 25%. By fully integrating the strengths of both parties and jointly expanding cloud infrastructure for the AI era, our collaboration with Xiaomi and Kingsoft in the AI space continued to progress and develop.

Tao Zou: Finally, in terms of profitability, this quarter, our non-GAAP gross profit was RMB327 million, representing a year-over-year increase of 9.6%. NAMGAP EBITDA margin reached 16.2%, an increase of 14.3% points year-over-year, mainly attributable to the continued increase in the proportion of AI business revenue. However, we also witnessed that our profits have experienced some fluctuation quarter-over-quarter, while non-GAAP gross margin declined by 2.6 percentage points quarter-over-quarter to 16.6%. The decline in gross margin was mainly due to declined profit contribution from a lower proportion of enterprise cloud revenue, as well as the impact of front loaded investment of computing resources.

Tao Zou: The non-GAAP operating profit was impacted by the decline in gross profit, resulting in the loss of RMB55.8 million this quarter. Non-Gap Operating Margin was negative 2.8%, representing an improvement of 4.4 percentage points compared with a loss of 7.2 percentage points in the same period last year. and turned from a profit to a slight loss quarter over quarter.

Tao Zou: Despite quarter-over-quarter fluctuations in financial performance this quarter and the headwinds of both market pressure and supply chain uncertainties, We remained firmly on tracking with our long-term strategy and continued to move forward with confidence. We have strengthened our foundation in ecosystem cooperation, computing infrastructure deployment, and AI application, advancing the strategic layout of our company's overall AI cloud services.

Tao Zou: Next, I will introduce to you the performance of the first quarter of 2025. For Kingsoft Cloud, the income of this quarter is RMB1.35 billion. 同比增长14% AI業務作為重點的增長驅動力 This quarter, the net income has increased significantly to 5.3 billion yuan. 同比增长超200% increased by 11% accounted for 39% of the total revenue of Kingsoft Cloud and continued to lead the industry. We are talking about eco-customers. on the basis of a stable increase in the volume of large model customers. Internet Client Business Scene AI Application such as online education, online travel, and other business strategy needs have also broken through. In terms of group building, we focus on customer demand rhythm and efficient coordination.

Tao Zou: In addition, we also adopted a flexible fund cooperation model. to ensure the rapid development of bottom-level strategy supply and support for AI business.

Tao Zou: Now, let me walk you through the key business highlights for the first quarter of 2025. In public cloud space, revenue reached RMB $1.35 billion this quarter, representing a year-over-year increase of 14%. AI business as a key growth driver reported a significant increase in gross billings to RMB 525 million, up over 200% year-over-year and 11% quarter-over-quarter, accounting for 39% of public cloud revenue, continuing to lead the industry. Based on steady usage growth in ecosystem customers and foundation model customers, computing demands for AI applications in internet customer business scenarios, such as online education and online travel, also made breakthroughs.

Tao Zou: In the construction of clusters, we efficiently coordinated and responded quickly to the demand schedule of key customers, created benchmark case of delivering full-spectrum cloud services for large-scale clusters within the quarter. In addition, through flexible capital cooperation models, we ensured sufficient underlying computing power supply to support the rapid growth of our AI business. Inc.

Tao Zou: 本季度實現收入 6.2 billion yuan 同比增长5% 本季度,我們受到季節性驗收節奏放緩的影響。 Han Yueyun's income has fallen. In terms of industry, 公共服务领域 to promote the application of AI in the cloud and the national cloud. Kick it 24 hours every day and it'll keep you busy 24 hours a day. Kingsoft Cloud combines data processing, model analysis, model assessment, model quantification, etc. and provides a full-fledged, temporary AI platform for users. to help customers optimize their model performance in real business scenarios. 数字健康领域本季度开始,武汉市检验检查结果,互认共享平台建设。 Kingsoft Cloud has once again been tested in Jiangsu, Chongqing, and other cities. It has expanded from imaging to testing and inspection, and digitally expanded in the Hubei market.

Tao Zou: In enterprise cloud space, revenue reached RMB616 million this quarter, representing a year-over-year increase of 5%. Affected by seasonal slowdowns in project delivery and acceptance process, enterprise cloud revenue declined quarter over quarter. By industry in the public service sector, we are advancing the application of AI in public service clouds and the state owned asset. actively embracing the AI-driven trend. Kingsoft Cloud has built a rich set of model resources through open source model marketplace while providing a one-stop model tool chain that covers key process, including data processing, model fine-tuning, model evaluation, and model quantization. We remain committed to delivering full process, one-stop AI services to help customers deeply optimize model performance in their business scenarios.

Tao Zou: In healthcare sector, we initiated the construction of a platform for mutual recognition and sharing of test and examination results in Wuhan. Kingsoft Cloud's medical imaging cloud capabilities, which have been deployed in regions such as Jiangsu and Chongqing provinces, were once again validated. Our capabilities also expanded from imaging scenarios to test and examination scenarios and have been further replicated and extended to the entire Hubei province market.

Tao Zou: In terms of product technology, we will continue to maintain the technological advantage. focusing on the first-class customer experience of creating core products. This quarter, we will continue to improve the product capabilities of our calculators. Kingsoft Cloud is an AI development and deployment platform. We strive to provide high-efficiency, flexible, low-cost model training and inference services for enterprises. including the high-quality model interface of the Xiaomi model will further expand the ecosystem of the platform and assist customers in the natural language processing, multi-modal interaction, and smart decision-making scenarios to realize technical applications. In terms of product and technology, we uphold the principle of building success based on technology and innovation.

Tao Zou: focusing on delivering best-in-class customer experiences across our core product offer. This quarter, we continue to enhance the product capabilities of our Intelligent Cloud Computing Service. Our Xingliu training and inference platform as a one-stop AI development and deployment platform remains committed to providing enterprises with efficient, elastic, and cost-effective model training and inference services. The integration of high-quality models, including the one from Xiaomi, will further expand the platform's ecosystem capabilities and help customers apply AI technologies in scenarios such as natural language processing, multimodal interaction, and intelligent decision making.

Tao Zou: 总体而言, Our AI business continues to develop rapidly, and Cloud services enter a new development cycle. In the public cloud field, we have the ability to build AI tools in the Hang Shi infrastructure, training platform to help customers reduce training costs and improve the stability, convenience and efficiency of model development and optimization during the use process. Hangyuan Yun Linyu, EI, Shuji, Bangongwei Zhuashou In the development trend of enterprise AI plus scenarios 提供一段式模型解决方案与服务 In the future, we will continue to work closely with Xiaomi Jinshan Ecology.

Tao Zou: to fully understand and explore new AI opportunities and continue to create value for our customers, shareholders, employees, and other stakeholders.

Unknown Executive: Next, let's welcome CFO Harry to introduce a set of financial results. Thank you. Overall, our AI business continued to grow rapidly, and our cloud services have entered a new development cycle. In the public cloud space, we strengthened our capabilities in infrastructure, training and inference platforms, and AI tools to help customers reduce training costs and improve stability, convenience, and efficiency throughout model development, fine-tuning, and usage. In the enterprise cloud space, with a focus on AI, data, and office productivity, we provide one-stop model solutions and services under the trend of enterprise AI plus scenario development.

Unknown Executive: Looking ahead, we will maintain deep cooperation with the Xiaomi and Kingsoft ecosystem, fully understand and explore new AI opportunities, and continue to create value for our customers, shareholders, employees, and other stakeholders.

Henry: I will now pass the call to our CFO, Henry, to go over our financials for the first quarter of 2025. Thank you. Thank you very much, Chairman Zou. Thank you, Mr. Zhao, and thank you all for joining the call today. I will now walk you through the financial results for the first quarter of 2025.

Henry: This quarter, our AI strategy continue to drive our growth and the laid a foundation for future development. Total revenues for this quarter were 1,970.0 million RMB, reflecting an 11% year-over-year increase. Out of this, revenues from public cloud services were 1,353.5 million RMB. up 14% from 1187.4 million RMB in the same quarter last year. This growth was mainly fueled by a surge in AI related business with a filling reaching 525 minute hour.

Henry: This quarter, our capital expenditure reached 605 million RMB. Revenues from enterprise cloud services reached 616.5 million RMB, up 5% from 588.2 million RMB in the same quarter last year. primarily driven by increased demand in industry solutions. However, we have witnessed a 25% sequential decrease of enterprise cloud revenues, which was mainly due to the seasonality impact. Total cost of revenue was 1,651.7 million RMB, up 11% year-over-year, which was in line with our revenue expansion. IDC costs dropped by 6% year-over-year from 768.5 million RMB to 722.8 million RMB this quarter, reflecting our execution on cost control and a better rack utilization.

Henry: Depreciation and amortization costs increased from 183.5 million RMB in the same period of last year to 378.5 million RMB this quarter, mainly due to the depreciation of newly acquired Solution development and the service cost rose by 13.3% year-over-year from 446.0 million RMB to 505.2 million RMB. driven by expansion in Camelot personnel to support revenue growth.

Henry: Fulfillment cost and other costs were 3.1 million RMB and 42.1 million RMB this quarter respectively. Our adjusted gross profit for the quarter was 327.7 million RMB, a 9.6% increase year-over-year, while a decrease of 23.4% quarter-over-quarter. Adjusted growth margin was 16.6% in this quarter compared with 16.8% in the first quarter of 2024 and 19.2% in the fourth quarter last year. Our adjusted growth margin has been negatively impacted by the seasonality for enterprise cloud service. and higher upfront investments into servers and RACs for AI business. On the expenses side, excluding share-based compensation, our total adjusted operating expenses were 427.3 million RMB, a decrease of 9% year-over-year, and a 4.3% quarter-over-quarter, of which our adjusted RMD expenses were 200.8 million RMB, increased by 4% from the same quarter last year.

Henry: Adjusted selling and marketing expenses were 107.8 million RMB, increased by 10.1% year-over-year. Adjusted G&A expenses were 118.7 million RMB, decreased significantly by 13.6% year-over-year due to the decline of credit loss. Our adjusted operating loss was 55.8 million RMB, narrowed by 56% from 127.0 million RMB in the same period of last year. The improvement was mainly due to increase of gross profit and a decrease of credit loss expenses. However, the adjusted operating profit turned loss from last quarter, which is mainly due to the decrease of gross profit in this quarter. Our non-gap EBITDA profit was 318.5 million RMB, increased by 8.3% of 33.2 million RMB in the same quarter of last year.

Henry: Our non-gap EBITDA margin achieved 16.2%, compared with 1.9% in the same quarter last year. It was mainly due to our strong commitment to AI cloud computing development, strategic adjustment, business structure, and our strict control over cost and expenses.

Henry: As of March 31, 2025, our cash and cash equivalents totaled 2,322.7 million RMB, providing a strong liquidity position to support operations and in AI investment. Looking ahead, our cloud infrastructure is ready to serve in a short time. The demand from ecosystems and other AI application scenarios not only fueled our business growth, but also reinforced our confidence in this trajectory. With demand for AI cloud service continues to grow, we are well positioned to capture and capitalize on AI capital opportunities.

Unknown Executive: Thank you.

Unknown Executive: This concludes our prepared remarks. Thanks for your attention.

Unknown Executive: We are now happy to take your questions. Please ask your question in both Mandarin Chinese and English, if possible.

Unknown Executive: Operator, please go ahead. Thank you.

Unknown Executive: As a reminder, to ask a question, please press star 11 on your telephone and wait for your name to be announced. To withdraw your question, please press star 11 again.

Unknown Executive: We will now take the first question.

Brian Gong: from the line of Brian Gong from Citi. Please go ahead. I have two questions. The first question is, in the first quarter, we saw that the growth of public cloud and industry cloud seemed to be a little weaker than we thought. What is the reason behind this? And how should we look at the revenue situation for the whole year? The other question is, Xiaomi has also released its own model. Can you please introduce the latest requirements of Xiaomi for us? What changes have we seen? Thank you.

Brian Gong: I will transmit it myself, thanks to management. So one is in the third quarter for both public cloud and the enterprise cloud, the growth seems a little bit weaker than our previous expectation. What are reasons behind this? And how should we see the pull year growth right now? And secondly, recently, Xiaomi released its own last language model. Can management elaborate what's the latest demand from Xiaomi? Thank you.

Tao Zou: I'll answer first, and then we'll see what Harry and Liu Tao have to say. The first question is about Q1's public and private users and the industry. We actually talked about this at the conference. It's true that it's affected by the season. It's still a big impact, whether it's our headquarters or Crypto. Yes, because Q1 has a Spring Festival. And a lot of industry customers, regular customers, in the whole of Q1, relatively speaking, Xiaodan Zhang, Linlin Yang, Dapeng Gong, Wenting Yu, Kingsoft Cloud As for Kingsoft Cloud, I can tell you that Liu Tao can make up for it.

Tao Zou: Because we mainly focus on big customers. Basically, there is a cycle of construction. The cost will gradually be visible until the next Q. For example, in Q1, we actually paid a 502k of Xiaomi. It's basically related to your second question. But in fact, we're basically at the end of the Q1 quarter. It's basically at the end of March. So the entire income can only be reflected in the second quarter. So, I used to talk about games. I often talk about the staircase curve. Because we focus on big companies. Basically, it may take a quarter to build.

Tao Zou: So, in this quarter, you see a lot of investment. The impact of income and profit will be reflected in the next quarter. So, by the next quarter, you'll see that the supply chain will be faster. So, I think this is about your first question.

Tao Zou: The second question is... 关于小米模型这个对我们的影响,其实刚才已经谈到了,实际上呢, It was released in May. No, it was released in April. The GB model was basically what came out of our group after we switched. Yes. So, and... As we just said, the income increase of QE in Xiaomi's ecological system is only 50% compared to last year. In fact, we haven't disclosed AI alone yet. Actually, AI is... So, in general, this also reflects that the data is compatible with Xiaomi's acceleration in AI itself, okay? From a long-term perspective, The scope of imagination is even wider. Whether it's in the training of the model, or as Xiaomi's model becomes more mature, it applies to its wide ecological product range, it brings a huge demand for reasoning.

Tao Zou: I think there is still a greater scope of imagination in the future. Of course, all of this is because we are serving Xiaomi after all. and Tenzuya Tanabata. and the development of Xiaomi itself. 这个接踵而走的,好不好? Thank you very much for your questions.

Tao Zou: So in relation to the first question about the declined speed in terms of both public cloud and enterprise cloud, as we noted in the prepared remarks, especially for the enterprise cloud, the seasonality is quite evident. That includes the impact for both our own enterprise cloud services as well as the business from Camelot, because obviously Q1, there's the factor of the Chinese New Year and a lot of customers for enterprise cloud business are still doing their budgeting. Now in terms of the public cloud, as we mentioned before, the public cloud business of Kingsoft Cloud focuses mainly on key customers, which typically are large customers, and they have their own cycle of the construction of the clusters before they can actually get online and to generate revenue.

Tao Zou: So in fact, in the first quarter, we have delivered a 512-node cluster to our key customers. However, that delivery time point was only at the end of March, and therefore the revenue and profit reflection as in the financial numbers will only be shown in the next quarter, which is the second quarter Q2. Now in relation to your question about the model from Xiaomi, in fact, the Xiaomi 7 billion parameter model was actually trained from our cluster, and in fact, the tremendous growth of the KC's AI business has much to do with the demand coming from Xiaomi.

Tao Zou: In the long-term perspective, the tremendous demand for model inference coming from Xiaomi still has significant potential. However, since we're serving Xiaomi, the pace of Xiaomi's demand has a lot to do with when and the amount the revenue and profit will show on our financial statement on a quarterly basis.

Unknown Executive: Last question, please. Thank you.

Unknown Executive: We will now take the next question.

Xiaodan Zhang: From the line of Xiaodan Zhang from CICC, please go ahead. Good evening, Chairman Zou, Henry, Clark, and Nicole. Thank you for accepting my question. My first question is that I noticed that recently Kingshan Cloud and Kingshan Office jointly launched an AI integrated machine in the business field. I would like to ask the management team to share how they view the business opportunities in the business AI field. Can you also introduce the pricing model of this AI integrated machine and the model of interest allocation between us and the office?

Xiaodan Zhang: The second question is about our profit margin. We can see that the non-gapped OP margin has declined since the last quarter. In fact, this quarter, there has been a decline in the profit margin due to the impact of the season. How does the management team view the trend of profit margin in the following quarter? Let me quickly translate my question.

Xiaodan Zhang: So thanks management for taking my questions. And first of all, I noticed that the Kingsoft Cloud and Kingsoft Office have recently jointly launched an AI all-in-one model machine for government affairs sector. So how do you view the business opportunities in the government affairs AI field? And also could you introduce the pricing models of this all-in-one machine? And secondly, could you please update on the non-gapped OP margin outlook for the subsequent quarters? Thank you. Okay. Thank you Xiaodan.

Unknown Executive: Question number one, the person in charge of gaming today, Ruigrong from our Zhengxi Team, at the scene. So I'll ask him to answer.

Unknown Executive: I'll be back in a minute.

Unknown Executive: The second question will be answered by Henry later.

Unknown Executive: Hello, I'd like to give a brief explanation on the issue of government-to-government collaboration. The government-to-government collaboration is a very fast-growing area. So, we and our partner, Jinshan Office, are working on a solution for government-to-government collaboration in some important cities in China. This solution is also a stage-by-stage market verification. It is also a very important demand for future development.

Unknown Executive: What is an integrated machine? It is a solution that combines heavy-duty solutions and AI computing hardware. It is small, flexible, and suitable for various types of customers. So, last month, we worked with Jinshan Office to develop a reasoning scenario for the development of office models in this industry. The software and hardware that are released are integrated products. This market is mainly targeted at customers from government-to-government branches in China. So, this is the general situation of an integrated machine. As for the question you just asked about pricing, it depends on different hardware. So, our integrated product covers basic configuration products, including some domestic hardware products.

Unknown Executive: So, the price is different. So, it is also gradually being promoted in the market.

Unknown Executive: That's basically all I have to say. So let me quickly translate. So as Riley pointed out, the market for AI in the office automation targeting the government space has been developing very fast and therefore that's the reason why Kingsoft Cloud and come together with Kingsoft Office to jointly develop and offer a plan or a solution to such market, which has been actually validated by the market. So that integrated solution is actually an integration of both software and hardware, and notably in the software portion, there's the AI component that helps them to become more efficient in office scenarios.

Unknown Executive: And this is targeted for Chinese government agencies in various layers, various tiers. Now, in terms of the pricing question you asked, it actually varies based on the different configuration of hardware that we are providing to the customer. So the pricing can actually vary by quite a lot.

Henry: Thank you for the question on the margin side. I think probably a few things I want to point out. So, you know, given the seasonality in Q1, as probably many of you understand that, you know, our primarily OPEX is the human capital expenses. So the salaries, compensation, bonuses, and so forth. So in Q1, basically, we not only need to incur the normal salary payment, but also there's a bonus and also certain kind of employee benefits also incurred in Q1 during the holiday seasons, as you understand. Given the revenue side was affected by, you know, the delay of certain projects, and also the top line growth has been affected by seasonality.

Henry: So the variation of the top line growth largely affected the bottom line. So the cost structure, especially on the salary and the compensation are relatively fixed. So that's why it's kind of affecting on the margin Q1, it's kind of very straight line on the reason. So going forward, I think given the. better margin profile. For example, the AI projects and our relationship with our ecosystem partners, especially for Xiaomi as our CEO, Mr. Zhao just mentioned, the big projects were likely to deploy on the way and it will be largely a book starting from Q2 going forward.

Henry: So hopefully our top line growth will drive a better margin expansion in the following quarters. So we are in the view that our OP margin will be better in the following quarters. It depends on the top line growth, the pacing of that, and the scale of the revenue quality in the following quarters. But on the other end, we also want to point out, given the EBITDA margin will be likely better compared with the OP margin, given the nature of the AI business. So as the AI business penetration becomes higher in the public cloud services, so hopefully the EBITDA margin can be recovered a little bit faster than the OP margin side.

Henry: So I think I will encourage the audience to probably closely track our growth margin profile and EBITDA margin profile, as well as the R&D cost, which are primarily linked to the expenses and the compensation and salaries of employees. I think these are factors that are going to be the leading indicator for the OP margin going forward for the following quarters.

Henry: So while we don't give a formal guidance, I understand Brian also mentioned this, for the four-year top line revenue in the call today. By the way, I interviewed that margin profile in the following quarters, likely, especially in the second half of this year, will be better than the first half of this year. And at this moment, we don't give a formal management guidance for the top line, but the margin profile will continue to be better, especially in the second half of this year.

Unknown Executive: Thank you.

Unknown Executive: Thank you for watching.

Unknown Executive: Please subscribe to my channel. Thank you.

Unknown Executive: We will now take the next question.

Thomas Chong: from the line of Thomas, Chong from Jeffreys, please go ahead. Good evening. Thank you for accepting my question. My question is about the AI cutbacks and the cost. Can you elaborate on the breakdown in the past three years? Also, I want to ask more about the chip ban issue. versus Q1, will there be any impact? And the last question is also about some industry issues. Regarding people starting to use these small models, will there be any impact on our remote services?

Thomas Chong: Thanks, management, for taking my question. My first question is about the AI cutbacks and the uptakes breakdown. Can we have an update like last quarter? And my second question is about our quarterly cutbacks like Q2. Is the trip ban issue affect the sequential momentum in cutbacks?

Thomas Chong: And my third question is more about the industry landscape. Given that customers are now using more like a distilled model, smaller models, rather than larger models, would that affect the cloud revenue? Thank you.

Henry: Thank you, Thomas. I will take the first question and defer to CEO, Mr. Zou, for the second and third question. I will answer the first question first. The second question is about chip restrictions and small model influence.

Tao Zou: Mr. Zou, please go ahead. The first question regarding the capex expenditures. So I have mentioned in this quarter, our total capex was 605 million RMB for the first quarter of this year. So if the audience remembers, since last year, we actually diversified the way we've financing the investment into AI, especially infrastructure construction. So our own cash to be used for the Capex is one part, but also we are going to, for example, doing a financial leasing, operation leasing with our partners, for example, Xiaomi as well. But also we are going to get the financing, for example, the bank loans and a certain leasing agreement with the state-owned banks in China locally, as well as some leasing companies in China as well, to arrange certain off the balance sheet of financing arrangements.

Tao Zou: So I think these are the three ways that we're financing the total AI infrastructure demand. So out of that, I think the reason you probably point out the total Capex for this quarter is kind of 600 million for this quarter. But I think our total investment into AI, because we also start to leasing certain servers from third parties to reduce our burden to pay out of pocket from own cash. So there's a certain leasing and a rental agreement we actually start to negotiate and set up with third party stock from Q1. So hopefully for next quarter, we're going to give some updates regarding our total spending, including both our own cash out of pocket, but also the leasing agreement we'll start to arrange with third parties.

Henry: So this is going to be the total investment of the AI infrastructure from our definition. So we're going to likely have some updates for next quarter because we just start to arrange those agreements with third parties, ensure to reduce the saving and to be more efficient in terms of our own cash management. So I think right now we give the number on total Capex, but the leasing plus the Capex and plus our own cash will be another update for next quarter. And also given the OPEX, as I mentioned in earning costs, our R&D expenses for this quarter is around kind of 200 million RMB and the sales expenses is around 108.

Henry: Uh... million RMB. And the management SG&A is around 119 million RMB. So if you're putting the three numbers together, it's roughly about 400 million on the OPEX side, including R&D, self-marketing and SG&A. I think this probably can give you a kind of a color regarding the total OPEX plus CAPEX for on a quarterly basis.

Henry: Thank you.

Tao Zou: I'd like to invite CEO, Tao Zou, to talk about the issue of chip restrictions and small models. Let me answer the second question. I don't need to translate. Let me answer the second question. Guanyin this can be very effective 具体的影响我想展开来说一下 Dina 其实对于这种 It's not the first time, and it's not the first day, so we, uh, this In this wave, we were still prepared, and we also made appropriate preparations. So, in the short term, especially since we are mainly a big customer of service providers, in the short term, The impact is not great. But in the medium term, it will not only affect our industry.

Tao Zou: 都还会有比较大的影响,这是第一个结论啊, This is the first conclusion. The second one is that in 2018 This year we really appreciate 为了应对这种困难 为了解决这种困难 to make. I have downloaded the complete download for this project on my Periscopese and Guo Nei. were working closely with brand providers. I remember we had a Q&A session before and I said, actually, the model of our office today... welcome. 自己训的模型都是基于国泰算力去做的 We have been working on this cooperation for a long time. so from the long and the short view 必须要做好 国产替代的这种准备 We will continue to work hard and prepare for the future, okay? This is the second point.

Tao Zou: To put it simply, in the short term, so it will be effected for a long time and affect the investor more But we named this stock We have made a list of the changes 方案和可行性的准备 OK.

Tao Zou: OK, so very quickly to translate Mr. Zhao's response to the chip ban question. So honestly and very frankly, there surely will be impact, right? However, this is actually not the same time for similar kind of restrictions to be imposed on chips. So we have experienced similar things before. And so before this round of restriction actually hit us, we have had some inventory and storage. So therefore, the short term impact is not that material. However, from a long-term, from a mid-to-long-term perspective, there will be meaningful impact for both the Kingsoft Cloud as a company and for the industry as a whole as well.

Tao Zou: We mentioned before, since 2023, we have already strengthened our cooperation with Made in China Computing Resources. And that kind of cooperation, we have managed to continue that cooperation since then. And in light of similar restrictions, we have made, and we have been well-prepared for the gradual substitution of Made in China Computing Resources, should restrictions continue to be more restrictive.

Tao Zou: The third question is... Our business manager Liu Tao, the SVP of Kingsoft Cloud is here, so I asked him to come back. First of all, I'd like to make a conclusion. The small model, that is, the reasoning of customers using small models has no impact on the growth and income of our business. The reason is this. First, our traditional large model customers, their models are all top-notch. So they are basically all in the process of training and training. In addition, for our Internet customers, their main use of the model from the beginning is a smaller model or a medium model, such as this 30B to 10B.

Tao Zou: So this business is always a source of pure growth for us. In the future, the growth of the business of Xiaomi Ecology may be a positive thing for us. As Xiaomi's resource model, WPS, GDBC, and other open-source resource models grow in number, there will be more reasoning power to return to the ecology of Jinshan. It may have been transferred to other big model APIs, but now these amounts have returned to become their own reasoning power. So for us, it is an incentive for growth. So, the conclusion, the basic conclusion is that there's no negative impact for revenue and profitability for Kingsoft Cloud.

Tao Zou: And reasons being followed. The first one is for the traditional large-language model customers, their models have been 100 billion parameter large-language models, and it has historically been using our computing power for training and inferences, and this part is actually not impacted. And secondly, for the more traditional internet space customers, they have actually always been using relatively smaller models, including the 30B and 10B kind of parameter amounts for our business. So, this, for us, is actually an incremental amount for our revenue development. And thirdly, in terms of the future for the Xiaomi and Kingsoft ecosystem, we also expect to have actually positive impact coming from the more pervading usage of relatively small models, because more of the model inference was originally coming from outside of the Kingsoft Cloud computing power infrastructure from other companies.

Tao Zou: Actually, when the Xiaomi and Kingsoft ecosystem started to adopt the more medium-sized or smaller-sized models, they were actually starting to use more of Kingsoft Cloud's computing resources. So, that's actually another part of potential incremental revenue and profit for us.

Unknown Executive: Thank you all for joining our next Q&A. Thank you.

Wenting Yu: We will now take the next question from the line of Wenting Yu from CLSA. Please go ahead. Good evening, Chairman Zhou, Henry, Clark, and Nicole. I'm Wenting, an analyst at SAIC. Thank you for the opportunity. I have two questions to ask.

Wenting Yu: The first question is, can you share with us the recent trend of the interest rate of domestic AI cloud rental services? We have seen that the competitive price in the industry is mainly around the large model API price. Do we see some pressure on the price competition in the AI service rental services?

Wenting Yu: The second question is, after we saw DeepSight open R1, did we observe that the training needs of the industry have improved or slowed down? Because I understand that everyone is doing this post-training model. This part will have an increase in demand, but the opening of R1 may also allow some model factories to give up and continue to replace models. How do we see the impact of these two factors on the training needs of the entire industry?

Wenting Yu: So the first question is, can management share the recent gross margin transfer AI cloud leasing services? Previously, we observed the aggressive price competition in the industry, mainly focused on model API pricing. Are we seeing the pricing pressure extend to the AI server leasing business as well?

Wenting Yu: And the second question is, following the open sourcing of R1 by DeepSight, how do we see the latest dynamics of model training demand? While there is incremental demand for post-training models, the open source release of R1 may also lead to some model vendors abandon further iteration. So how should we evaluate these two factors and their overall impact in the industry? Thank you.

Tao Zou: Okay, I'll answer the first question first, and then Tao will answer the second one. The first question is about domestic competition. The Maoli trend I sat outside of the door where I got in the car Why Buddha? Cheers! As you said, it's the whole market. As AI players focus on the project Chen, Xiaodan Zhang, Linlin Yang, Dapeng Gong, Wenting Yu, Kingsoft Cloud This will have a certain impact. Especially our new projects, because the old ones... One for each. Hi, everyone. So this is to a certain extent, from a general point of view. Huiyun Lin, Yixiao Zhang, Xiaodan Zhang, Linlin Yang, Dapeng Gong, Wenting Yu, Kingsoft Cloud 当然影响的程度呢 And of course, the degree of impact Fengzi is small, Fengmu is large, so the impact is small.

Henry: If the volume is large... Thank you for watching. Bye bye. This whole thing will have a bigger impact, this is one. The second one is, in general, We call it the resource pool supply method. 下面陆续开始吧,我们会... You can find me on Facebook at https://www.facebook.com and Twitter at https://www.twitter.com. and other companies in China. and some non-corporate clients. So, to a certain extent, this will also have a certain impact on our毛利. To put it simply, it's like a split-day model. They will be doing thisäh? It will also have a certain influence on the overall usage in China this year.

Henry: So, in relation to the AI cloud service margin or pricing pressure question, let me respond to you from two dimensions, which is internal and external. So, externally, as you rightly pointed out, we're actually seeing sort of market concentration for AI players as our customers. And therefore, they do have some impact to our new projects, which we entered into more recently. And the impact would largely depend on the project size. However, for the old long term contracts that we have signed in the past and for those projects that we entered into in the past, there's no meaningful or no material impact.

Henry: Another point I want to point out is that starting from the second half of 2024, we're increasingly leveraging the so-called resource pool supply chain to expand our infrastructure. So, the strategy basically is for top of CA customers, we use our own cash and dry powder and cutbacks to build the infrastructure and to provide cloud service to them. However, for non-top customers, we're increasingly leveraging the partners who will supply us the servers and computing resources. And we're jointly together to provide cloud service to these customers. And because these are partners that provide resources, we actually also have to share some of the profit with them.

Henry: So, this is like a partnership in profit sharing model. And for that reason, that new or emerging supply chain model has also, to some extent, negatively impacted the gross margin level.

Henry: The second question is, from the point of view of training needs, we can see that because of DeepSake's emergence, the training needs of domestic large model customers are increasing. have begun to have the need to train large models themselves. So from this point of view, we think that the overall training needs of large models are relatively stable. On the one hand, for example, large model startup companies may have a certain amount of training. But for new players, they may think that we have the right resources to train a SOTA that they can control. So we think that the overall demand is stable and slightly increased.

Henry: So in response to your question regarding the impact of open source of DeepStick R1 to the demand of model training, my answer to you is this. This is coming from our SVP, Mr. Liu Tao. So the old batch of large-language model companies, the so-called independent large-language model companies, we do have seen their demand shrinking to some extent due to the success of DeepStick. However, most of the contracts that we entered with them are long-term contracts, which we have entered in the past. So as we responded in the first question, so the impact is very limited.

Henry: However, there's also a group of other companies other than those large-language model companies, some of them coming from the internet space, some from other emerging industries. They are inspired by the success of DeepStick. And also think that by using relatively manageable resources, they will also potentially be able to train and owe a state-of-the-art model. So therefore, we're seeing increasing demand coming from these customers as well. So all in all, we don't see negative impact from the success of the DeepStick model. Yeah. OK. What other reasons? Right there? Thank you.

Unknown Executive: There are no further questions at this time. This concludes today's conference call. Thank you for participating. You may now disconnect. Thank you for watching!