XPeng Inc.

Hello, ladies and gentlemen. Thank you for standing by for the third quarter 2024 earnings conference call for XPeng Inc. At this time, all participants are in listen-only mode. After management's remarks, there will be a question and answer session. Today's conference call is being recorded. I'll now turn the call over to your host, Mr. Alex Shear, Head of Investor Relations and Capital Markets of the company. Please go ahead, Alex.

Thank you. Hello, everyone, and welcome to EXPON's third quarter 2024 earnings conference call. Our financial and operating results were issued via newswire services earlier today and available online. You can also view the earnings press release by visiting the IR section of our website at ir.xiaopeng.com. Participants on today's call from our management will include our co-founder, chairman and CEO, Mr. He Xiaopeng, vice chairman and president, Dr. Brian Gu, Vice President of Corporate Finance and VW Projects, Mr. Charles Zhang, Vice President of Finance Accounting, Mr. James Wu, and myself. Management will begin with prepared remarks, and the call will conclude with a Q&A session. A webcast replay of this conference call will be available on the IR section of our website. Before we continue, please note that today's discussion will contain forward-looking statements made under the Safe Harbor provisions of the US Private Securities Litigation Reform Act of 1995. Forward-looking statements involve inherent risks and uncertainties. As such, the company's results may be materially different from the views expressed today. Further information regarding these and other risks and uncertainties is included in the relevant public filings of the company as filed with the US Securities and Exchange Commission. The company does not assume any obligation to update any forward-looking statements except as required under applicable law. Please also note that EXFON's earnings press release and this conference call includes the disclosure of unaudited GAAP financial measures as well as unaudited non-GAAP financial measures. EXFON's earnings press release contains a reconciliation of the unaudited non-GAAP measures to the unaudited GAAP measures. I will now turn the call over to our co-founder, chairman, and CEO, Mr. He Xiaopeng. Please go ahead.

Hello, everyone. I'm He Xiaopeng.

Hello, everyone. I'm He Xiaopeng.

Hello, everyone. I'm He Xiaopeng. Hello, everyone. I'm He Xiaopeng. In September, our exchange rate broke 20,000 units and reached the best level in history. Not only that, but with the full-scale line-up and growth of our technical base, our profit margin in the third quarter has increased to 15.3%, which has created a new high in the history of Xiaotong and achieved continuous improvement in the five consecutive seasons.

In the third quarter of 2024, we surpassed our key performance targets. We delivered 46,533 units in the third quarter, reflecting a 54% increase quarter-over-quarter and a 16% increase year-over-year, beating the high end of our prior quarterly guidance. September's deliveries exceeded 20,000 units, marking a record high. Furthermore, thanks to the technology-driven cost reduction and growth in scale, our gross profit margin increased to 15.3% in the third quarter, achieving our pinnacle level and demonstrating continuous improvement for five consecutive quarters.

The crisis is often a turning point. The victory will make the entire enterprise stronger than it was before, so that we can accelerate our progress and realize our goal in the new era. I would like to thank all the shareholders and all the people who have supported Xiaodong in the past two years. In the past two years, our strategy and products management and organization have achieved and carried out comprehensive changes. And we are constantly supplementing the short version of various aspects such as marketing, channel design, etc. At the same time, Xiaopeng's firm investment in AI technology is beginning to turn into a better product experience, a better cost advantage. It makes our long version longer. Then we also surround customer orientation, management orientation, and overall thinking. We are building Xiaopeng's XPD, which is from product development, testing, supply chain, to pre-sale, listing, delivery, and so on. To make it more consistent, to ensure the future of the scale and system, to be able to continue to create a product, and then a product, to bring a bigger surprise to users, the star package.

Over the past two years, Xpeng has undergone significant transformation amidst challenges. However, I remain calm at the center of the storm because crises often present opportunities. Having overcome these adversities, Xpeng has emerged stronger than ever. We're now poised to accelerate our growth and move forward steadily. I would like to express my gratitude to all of our shareholders and everyone who has consistently supported us. We have implemented comprehensive changes in our strategies, products, management, and organizational structure. We've also addressed previous areas of improvement in marketing, sales channels, and design. Moreover, our firm investment in AI technology has begun to yield advantages in both product experience and cost efficiency, helping to bolster our competitive edge. By prioritizing customer needs, maintaining a business-oriented approach, and keeping the full picture in mind, we have established a robust capability across our entire operations, from product definition and research and development to pre-sale activities, product launches, and delivery. As a result, we have created a series of standout products that truly surprise and delight our users.

从今年开始已经是小鹏的第二个十年的创业。 我也相信未来的十年一定是AI的时代。 所以我会带领小鹏汽车 to be a firm effort to face the global AI technology industry, and to lead the big-scale applications of AI in the digital field. From the perspective of the development pattern of the entire industry, I think that in the next three years, that is, from 2025 to 2027, it will be the stage of elimination of the Chinese automotive industry. On the one hand, I think that the penetration rate of Chinese new energy vehicles will increase by more than 85%, On the other hand, I also believe that the change in AI will drive the entire market style to move to the next stage of temporary integration. In the past, we have seen that most of the car development of traditional cars is dependent on the cooperation and integration of supply chains. And in the field of AI cars, there must be depth and a wide range of full-time self-reliance capabilities. It will become the champion of industry competition. We will use AI's high-intensity computing power as a data engine to integrate cars, including AD, chassis, chassis, power, and other various hardware and software development, and upgrade them at a faster speed. This will allow us to experience faster and more powerful optimization, and also allow companies with larger and more traditional development models to take advantage of it. So I believe that from 2015 onward, children will see a huge difference, whether it is in automatic driving or in the whole car. I believe that AI's large model will bring multiple angles of subversive experience and change. It will allow more users to observe from the original to take the initiative to embrace a more secure, comfortable, and fully smart AI system.

Xpeng's second decade has started, and I believe the next 10 years will be the era of AI or artificial intelligence. I'll strive to lead Xpeng to become a global AI-defined car company and spearhead the large-scale application of AI in the mobility industry. Looking at the industry landscape, I anticipate that between 2025 and 2027, which is the upcoming three years, we'll see a knockout phase in the Chinese automobile industry. The penetration rate of China's new energy vehicles will likely rise to over 85%, while the integration of AI will lead to the next stage of consolidation of market share. Unlike the traditional car companies that have relied on cooperative integrated supply chain models for research and development in the past, winners in the AI-defined car sector will be those with in-depth full-stack self-development capabilities. We plan to harness the power of AI and use it as a data engine, integrated both software and hardware in our research and development of the whole vehicle, AD, cabin, and engines, etc. This will allow us to iterate and upgrade at an unprecedented speed, creating a substantial advantage over companies that continue to use traditional R&D models regarding user experience and optimization speed. Starting next year, I expect significant advancement in autonomous driving and vehicle intelligence. Introducing AI large models will provide a transformative experience from all aspects, enabling users to embrace safer, more comfortable, more comprehensive, and smarter AI driving and AI-driven vehicles.

On November 7, our world's first AI car, P7J, was officially launched. At that time, our total sales volume exceeded 30,000, and it is still in full swing. So the PG-plus has become a very obvious phenomenon in the medium-sized pure electric car market. I believe this is also an all-round milestone for AI cars. I am excited to see that the high-end car market is the core reason why users choose PG-plus. We all think that the past luxury is the luxury of configuration. I think that more and more cars are moving towards the luxury of technology. This is the direction Xiao Peng is heading. We are expanding the production capacity of PGJ with the suppliers. I expect that the delivery capacity of PGJ will reach more than 10,000 in December this year. Currently, M03 and PGJ have already launched a double production mode. I believe that as the production capacity rises, in the fourth quarter, the monthly payment volume will rise again. Of course, we will first work hard to have more than 30,000 monthly payments in November. We all know that the month of March is one of Xiaobong's main opportunities in 2024 and 2025. In 2025, we will bring tens of thousands of orders into the first quarter of 2025. So I believe this will make our payment volume more stable in the first quarter of next year.

On November 7, the world's first AI-defined car, the P7 Plus, was officially launched. That night, the number of firm orders exceeded 30,000 and continued to rise. The P7 Plus has become a phenomenal success in the mid-to-large BEV sedan market and marks a milestone in the widespread adoption of AI-defined cars. I'm pleased to see that the core reasons users choose the P7 Plus is its standard, high standard intelligent driving features across the entire model range. In the past, luxury was defined by configuration, but now it is defined by technology, and this trend is where we are striving to be. We're collaborating with suppliers to expand the production capacity of the P7 Plus, and I expect the delivery volume of the P7 Plus to exceed 10,000 units in December. Currently, both the M03 and the P7 Plus have begun double-shift production. As production capacity ramps up, we anticipate monthly deliveries will set a record in the fourth quarter and will strive to exceed 30,000 units in November. Additionally, steady and long-term progress is one of our main themes for 2025. will enter 2025 with tens of thousands of orders, which will increase our delivery volume in the first quarter of next year and lay a solid foundation for a significant increase in sales next year.

从P7加开始,那么我们之后的全新的MAX版本车型和改款都会采用国内唯一不依赖高清地图和金融内达的AI应用NGP自家方案。 In the near future, we will also bring it to the global market. I believe that in the global automotive industry, Xiaopeng will first use high-end self-priced software hardware to become the flagship of the whole industry, and make the experience a priority, rather than just in the promotion slogan. We hope to achieve the experience of level 3 with the cost of level 2, and to achieve the same price with technology and software. Then in the first half of next year's Spring Festival, with MOLA M03 MAX, it will also be mass-produced and the platform-based NGP project will be launched. We will be the first in the world to lower the price of high-end cars to the level of 150,000, which is the level of 20,000 US dollars. So the small and medium-sized cars will have the leading AI pricing capabilities and the ultimate cost control capabilities at the same time. I believe this is an unbeatable advantage for competitors.

Starting from the P7 Plus model, our new and facelifted MAX models will all feature the AI Hawkeye Visual ADAS solution. This is the only ADAS solution in China that does not rely on HD maps or LiDAR. And we'll take it to the rest of the world soon. In the global automotive industry, we're the first to standardize high-level intelligent driving software and hardware across our entire lineup, providing an actual leading experience for the users. We delivered experience of the Level 3 ADAS driving at the cost of Level 2, achieving what we refer to as intelligence-for-all tech-powered driving with the same cost as fuel. Excitingly, in the first half of next year, after the Lunar New Year, Mona M03 Max will start mass-producing the platform-based AI Hawkeye Visual ADAS solution. This will allow us to become the world's first car company to offer advanced intelligent driving vehicles for just 150,000 RMB or about US$20,000, lowering the threshold. With our leading AI technology and strong cost control, we have a competitive moat that will serve as our ultimate weapons for navigating from a fiercely competitive red ocean to the opportunities of the blue ocean market.

In the next 1 to 5 years, AI intelligent deepening will increase rapidly with unlimited speed. AI vehicles, which have strong AI capabilities and strong automatic driving capabilities, will speed up the replacement of these cars that do not have such capabilities, including those that only have promotional capabilities but do not actually experience good cars, will also be replaced. So by 2025, Xiaofeng plans to launch at least four new cars, including the first super electric version of the new model. In addition, we will also launch many new models.

I believe that over the next one to five years, the penetration rate of smart features will significantly increase non-linearly. Our AI-defined vehicles, which incorporate powerful AI capabilities and autonomous driving features, We'll accelerate the replacements of cars that lack these technologies or cars that only claim to have these technologies. In 2025, we plan to launch at least four new models, including super electric vehicles, and we'll also update several existing models. Each of these new and facelift models will be very distinctive in their respective market segments, and I look forward to launching more top-selling models that users will love next year.

In the recent Xiaopeng AI Technology Day, we released the Kunpeng Super Electric System. What is super electric? It is the next generation of real estate. We believe that Xiaopeng's super electric will be our second largest growth engine with our next generation of pure electricity technology. It will drive our acceleration development with our AI engine to a new level of global rapid development. So we know that power technology and energy leading has always been the label of Xiaopeng in the charging brand. We have a lot of users who praise us as the network of anti-fragile energy management. In the same way, in the next generation of our super power, we will also use the high-pressure charging technology of the leading market generation to solve the current user pain points of the current normal real-time solution and improve the user experience. So the super power of Kunpeng will use the third generation of 800V high-voltage system to do 430 kilometers of pure battery life, to do 1400 kilometers of comprehensive battery life, 5C super fast charge, and a lot of features, and also maintain excellent cost control. I believe this will lead to the change of the next generation of true technology. So in the future, Xiaopeng will adopt a two-wheeled way, facing the world, At our recent AI Tech Day, we unveiled the Kunpeng Super Electric System. Our next-generation extended range products, along with our next-generation pure electric products, will be our second-largest growth engine.

Together with our AI capabilities, it will drive strong momentum for accelerated development. Powertrain technology and industry-leading energy efficiency are key components of XPeng's brand. We have received widespread user acclaim for our exceptional energy consumption management with our electric vehicles regularly exceeding advertised range estimates. In our brand-new extended range products, we'll employ high-voltage electric technology one generation ahead of the market to address common user pain points facing current extended range products in the market and provide a user experience far superior to many existing extended range products. The Kunpeng Super Electric Drive System is built on our third-generation industry-leading 800-volt platform, supporting various features, including a pure electric range of 430 kilometers, a combined range of 1,400 kilometers, and the 5C ultra-charging battery, while also controlling costs, all of which will lead the way for the next generation extended range technology. In the coming future, Xpeng will adopt a dual energy approach, offering a batch of new vehicle models with pure electric and super electric powertrain options to cater to the diverse needs of global customers. I believe that this will significantly expand our total addressable market, bringing multiple opportunities for sales growth and accelerating the mass adoption of AI-defined vehicles worldwide.

In terms of business, PG&E will become the new starting point for Xiaohong's new generation of models to improve overall efficiency. The new models and important changes in 2025 and 2026 In terms of operations,

The P7 Plus will also mark a brand new starting point for the overall improvement in vehicle gross margins for Xpeng's next-generation models. New platform-based technologies we have implemented in the P7 Plus will also be applied to new models and major facelifts over the next two years of 2025 and 2026. We anticipate that the gross margin of our next-generation models will reach double digits, significantly increasing our sales volume to a new level during our strong product cycles and helping us move certainly toward achieving scale profitability.

Let's talk about AI pricing. Xiaotong's Turing AI pricing. We announced and realized from cloud to car, from software to hardware, even to chip, the full-time word of mouth. I believe this will lead to the next generation of full-time word of mouth technology. I saw on the market, A lot of friends are staying in the last generation of Xiaokong's large model architecture. I think the future ability of self-driving is in the cloud, in the data, in the algorithm, etc. Xiaokong's large model of cloud will be 80 times the speed of the car. I think this represents the leading level of self-driving in China. It makes the top and bottom lines of automatic driving Excellent, Turing.

AI smart driving systems demonstrate our robust full-set self-development capabilities. It integrates cloud-based and in-vehicle software and hardware, including chips. It sets a new gold standard for next-generation full-set development and highlights our exceptionally efficient R&D iteration process. Many of our peers are still using our previous generation of architecture and technological routes. I believe that enhancing the capabilities of smart driving relies heavily on cloud technologies. Our cloud-based large model has 80 times more parameters than the in-vehicle model, making it the most advanced technology currently available in China's ADAS market. In the coming years, the synergy between our in-vehicle data, cloud-based computing power, and both cloud and in-vehicle large models, and our globalization and car manufacturing will grow exponentially, marking and making significant leaps forward in our large model's performance per the scaling law.

At the end of this year, on the 5.5 version of Tianji, we plan to use a set of end-to-end NGP to open the whole scene from the car seat to the car seat. This will allow the overall experience of self-driving to begin to penetrate the new stage. In fact, three years ago, Xiaopeng was already able to basically open the car seat to the car seat. But that's when we used different rules and abilities to connect in different scenes. But from Xiaopeng's 5.5 is still a set. So I believe that 5.5 will allow Xiaopeng's frame to reach the original first tier, the standard between the non-neutral and the neutral. For example, it is between 50 to 60 points. It will exceed the original 50 to 60 points to reach the real full-fledged first tier. I believe that Xiaopeng's plan is to achieve a similar frame experience in the current hardware in 2025. That is, we hope to achieve 100 km efficiency control in about one time. We are currently developing a more advanced version of the Ocea model, which will greatly improve the car's performance, and add all-in-one design to the core whole car component system, making it more likely to be pre-installed and mass-produced RoboTaxi with low and safe costs. I believe that a huge improvement in the automatic driving capability will make AI capabilities In the future, it will accelerate into the core capabilities of a head-to-head vehicle, and it will lead the user's mind. At that time, users will find that AI does not only represent autonomous driving, but will also bring huge differences in AI assistants, AI hardwares, AI chassis, AI sound, AI battery doctors, and many other AI capabilities.

We plan to realize door-to-door full scenario ADAS on Tianji AI XOS 5.5 by the end of this year. This uninterrupted and ultra-smooth driving experience will elevate us from being on par with the first tier to truly leading the pack. We actually were able to deliver similar experience three years ago already, but it was a combination of multiple solutions, whereas this 5.5 OS is one-stop solution. Next, we plan to achieve a L3-like intelligent driving experience by the fourth quarter of 2025, targeting less than one takeover per 100 kilometers. The more advanced ultra version of the vehicle we're developing now will significantly enhance the computing power on board and incorporate a fully redundant design for core components. This will enable us to mass produce robotaxi at a low cost while ensuring sufficient safety. I firmly believe that the substantial improvement in autonomous driving capabilities will make AI a core differentiator among leading automating companies and a key factor in capturing user mindshare. Users will discover that AI is not only applied to autonomous driving, but will also expand and integrate into various aspects, including in-car AI systems, AI cabin, AR hub, smart chassis, smart audio systems, and AI battery doctors. In the medium to long term, the gap between AI leaders and laggards on product technology, brand image, and profit models will continue to widen.

Let's talk about globalization. In the third quarter, XiaoKong's overseas sales network has expanded to 30 countries and more than 110 stores, and we have achieved good opening and closing in many countries. XiaoKong is ranked first in China for its high-end pure electric models. Our XiaoKong G9 is also ranked first in North Europe for its medium-sized pure electric SUVs. In the third quarter, Our overseas sales overall increased by 70% and accounted for 15% of the company's sales. In 2025, we plan to further expand our sales network to more than 90% of the new energy market outside of North America and more than 300 stores. Our goal is to maintain high-speed growth in overseas sales in the next three years. Now let's talk about globalization. We're accelerating our global presence, leading the way for Chinese smart EV brands in their overseas ventures.

Our organizational management, product planning, autonomous driving technology, smart cockpit design, supply chain management, and manufacturing and production are all strategically aligned for global deployment. By collaborating with high-quality overseas dealers, we have extended our reach to more than 30 countries, with over 110 cell stores as of the third quarter. And we have experienced strong initial sales in multiple regions. Currently, XPeng ranks first in export sales of Chinese premium BEVs, and our G9 ranks first in the mid-to-large-size battery-electric SUV in northern Europe. In the third quarter, our overall overseas sales increased by 70% sequentially, accounting for 15% of our total sales volume. Looking ahead to 2025, we plan to further expand our international sales network to more than 300 stores, expanding to over 90% of the NDV market outside of North America. Our goal is to maintain robust growth in overseas sales over the next three years, aiming to secure the leading position in mid- to high-end NDV export sales among Chinese automakers. The rapid expansion of our international business will further boost our profitability.

After two years, I believe Xiaofeng is about to enter a new positive cycle. And we believe that in the fourth quarter of 2025, we will also begin to enter a faster-growing trend, accelerate growth and move towards profit. We expect that the total yield of the fourth quarter of 2024 will be about 87,000 to 91,000. Compared with an increase of 87% to 95.6%, The same rate rose from 44.6% to 51.3%. The revenue is expected to be 153 to 162 billion yuan. The return rate rose from 51.5% to 60.4%. The same rate rose from 17.2% to 24.1%. At the same time, we expect the fourth quarter, the cash flow will greatly improve to realize the free cash flow transfer in the second half of the year. The cash we have in our hands Up to two years of headwinds, we're about to enter a brand new positive cycle. In the fourth quarter of 2025,

we expect to experience tailwinds driven by AI transformation and the super electric system, which will accelerate our growth and lead us toward profitability. We anticipate our total delivery volume for the fourth quarter of 2024 will range from 87,000 to 91,000 units. This represents a quarter-over-quarter increase of 87% to 95.6%, and a year-over-year increase of 44.6% to 51.3%. Additionally, we project our total revenue for the fourth quarter to fall between 15.3 billion RMB and 16.2 billion RMB, reflecting a quarter-over-quarter rise of 51.5% to 60.4% and a year-over-year increase of 17.2% to 24.1%. Moreover, we expect our cash flow in the fourth quarter to improve significantly, resulting in positive free cash flow for the second half of the year. By year-end, we anticipate our cash on hand will exceed 40 billion RMB. With healthier gross profit and cash flow, we'll have the capacity to invest deeply in research and development for the future, allowing us to consistently and confidently provide our customers in China and abroad with market-leading AI-defined vehicles.

Thank you, everyone. Next, we have our financial VP, James, to introduce to you. He was born in 2014.

Thank you, everyone. With that, I'll turn the call over to our VP of Finance, James, to discuss our financial performance for the third quarter of 2024. Thank you, Xiaopeng.

Now let me provide a brief overview of our financial results for the third quarter of 2024. I'll reference RMB only in my discussion today unless otherwise stated. Our total revenues were $10.1 billion for the third quarter of 2024, an increase of 18.4% year over year, and an increase of 24.5% quarter over quarter. Revenues from vehicle sales were $8.8 billion for the third quarter of 2024, representing an increase of 12.1% year over year, and an increase of 29% quarter over quarter. The year over year and quarter over quarter increases were mainly attributable to higher deliveries. Revenues from services and others were $1.31 billion for the third quarter of 2024, representing an increase of 90.7% year-over-year and an increase of 1.1% quarter-for-quarter. The year-over-year increase was mainly attributable to the increased revenue from the technical R&D services related to the platform and software strategic technical collaboration, as well as electrical electric architecture, also known as EEA, technical collaboration with the Volkswagen Group. The quarter-over-quarter increase was mainly attributable to the revenue from technical R&D services related to the EEA technical collaboration with the Volkswagen Group, partially offset by the reduction in parts and accessory sales. Gross margin was 15.3% for the third quarter of 2024, compared with negative 2.7% for the same period of 2023 and 14% for the second quarter of 2024. Vehicle margin was 8.6% for the third quarter of 2024 compared with negative 6.1% for the same period of 2023 and 6.4% for the second quarter of 2024. The year-over-year increase was primarily attributable to the cost reduction and the improvement in product mix. The quarter-over-quarter increase was mainly attributable to the cost reduction. R&D expenses were $1.63 billion for the third quarter of 2024, representing an increase of 25.1% year-over-year and an increase of 11.3% quarter-over-quarter. The year-over-year and quarter-over-quarter increases were mainly due to higher expenses related to the development of new vehicle models as the company expanded its product portfolio to support future growth. SG&A expenses were $1.63 billion for the third quarter of 2024, representing a decrease of 3.5% year-over-year and an increase of 3.8% quarter-over-quarter. The year-over-year decrease was primarily due to lower employee compensation in the third quarter of 2024, while the quarter-over-quarter increase was mainly due to higher commissions paid to the franchise stores. As a result of the foregoing, loss from operations was $1.85 billion for the third quarter of 2024 compared with $3.16 billion for the same period of 2023 and $1.61 billion for the second quarter of 2024. Net loss was $1.81 billion for the third quarter of 2024 compared with $3.89 billion for the same period of 2023 and $1.28 billion for the second quarter of 2024. As of September 30, 2024, our company had cash and cash equivalents, restricted cash, short-term investments, and time deposits in total of $35.75 billion. To be mindful of the length of our earnings call, I would encourage listeners to refer to our earnings press release for more details on our third quarter 2024 financial results. This concludes our prepared remarks. We'll now open the call to questions. Operator, please go ahead.

Thank you. If you wish to ask a question, please press star 1 on your telephone and wait for your name to be announced. If you wish to cancel your request, please press star 2. If you're on a speakerphone, please pick up the handset to ask your question. For the benefit of all participants on today's call, if you wish to ask your question to management in Chinese, please immediately repeat your question in English. For the sake of clarity and order, please ask one question at a time. Management will respond, and then feel free to follow up with your next question. Your first question comes from Tim Hsiao with Morgan Stanley. Please go ahead.

I have two questions. The first question is about the price. This is what I just wanted to share with you. If the time is a little longer, I would like to ask the management level, in the next three to five years, do you think this difference will be enlarged or reduced? Because we also see that domestic electric vehicles are considered to be the standard for all types of vehicles. Although the price of the same enterprise is short-term, the price-to-price ratio is largely ahead, but once the price reaches a certain degree of scale and generalization, will the competition return to the cost of the competition mechanism? So my first question is about your time driving. Because in the next three to five years, are you expecting the technology gap of smart driving to be widened or narrowed? Several leading local EV brands in China are considering making smart driving a standard configuration for all mass market models. Will the time driving become a Me Too function in the future? And how could X-Pen ensure consumers can feel and appreciate the difference and choose X-Pen's cars because of that? That's my first question. Thank you.

Yes, thank you. That's a very good question. To be honest, in the past two years, we have been thinking about this problem a lot. First of all, I think that X-Pen's large model it will lead to higher price thresholds, because we need more money, more profit, more data to make it possible. In many areas, such as China and the world, these are all new thresholds. At the same time, I think that in three to five years, from our perspective, it will not only allow software and simple hardware to recover, We now have four areas of automatic driving, from the cloud to the car, from the chip to the battery, from the EEA to the robot, all need to go to fully automatic driving and cross the other people in the car. So the first one, I think, is to increase the range of fully automatic driving so that its ability will change. So this is the first change. I think its gap will be enlarged. The second one is that only in automatic driving, we think that in automatic driving, is the same as the age of the car. We believe that in the age of the car, there will be commercial use and real good use. There may be a hundred to tens of thousands of times the gap. This is in the next one to two years. I think this will be the second point of automatic driving change. The third point, of course, there will be a whole set of logic in the future, including the whole set of logic of a smarter car. So I think Three to five years will further increase the depth of impact on customer experience. It will make the head have a stronger scale effect.

Thank you. This is a very good question. Actually, we've been talking about this for the past two years while we are developing our end-to-end large model solution. Now, what we need here for this next generation of capability is not only, first of all, the capital for R&D, but also the computing power and big data as well. And in the coming three to five years, I think any companies who try to compete in this landscape will not only have the full-size self-developed R&D capability that combines software and hardware, but also on the cloud side, whole vehicle, chip development, and also EEA development across different car manufacturing capabilities are all essential for having that capability. So in sum, I believe that the gap between different EV makers will actually be widened in the several years. Another point that I would like to mention is that ADAS capability is just like part of your brain. You have to not only have your mindset, what you think you can do, but you actually need to be able to deliver your claims And that also will set the bar or the threshold for entering this competition. Another point that I would like to mention is the whole vehicle capability. How do we make sure that the whole car gets smarter in order to carry all this ADAS capability? So I think going forward, users will actually have a better idea or awareness of how capable a company is, how good the product is, and they will have an in-depth experience or first-hand experience of what truly means to have a smart ADAS capability in the product. Thank you.

Then I would like to add a small paragraph.

I'm sorry.

I would like to add a small paragraph. In the case that the original automatic driver can use Tier 1 and the entire car factory for PK, As I said, from the small to the big car, all the body parts can be driven by AI. It is impossible for TO1 to do all the AI. If it does, it will be a whole car factory. So, in the next three to four years, TO1 will have a rapid change in the entire industry. Thank you.

I also would like to add that traditionally the model for OEMs to develop cars is to work with Tier 1 suppliers. However, in the future, when we require cars to adapt the ADAS capability from not only the brain, but your upper tassels and your legs and the whole body, it will actually require a completely different model of development, which will also in the coming three to four years set us apart from the rest of the competition. Thank you.

My second question is about profitability. Because over the past few quarters, we've seen Expo posting consecutive margin improvement at both weekly and to the group level. Looking to next year, how could the company further narrow the loss and systematically turn the profit? That's my second question. Thank you.

Hey, Jim. It's Brian. Let me address this question. First of all, I think... In this quarter's financials, I think we're very encouraged to see that our non-GAAP operating margin, actually the loss has narrowed to 15.5% compared to in the second quarter, I think it's about 19%. So you start to see operating leverage, narrowing of our operating losses. And I think that trend will continue as we start. mentioned that we're going to launch P7+, which we think is actually a better margin profile product. And also, overall, we actually can see the scale effect come into play, as well as, additionally, we can see continued improvement on the vehicle margin front. And also, on the expense side, I think we also start to see very significant reduction opportunities For example, for the fourth quarter, I think we are going to hold a very consistent R&D spend below 2 billion R&D in total. And that will make our entire year's R&D to be probably less than 6.5 billion R&D, lower than our original estimate. So going into next year, we start to see these factors going to compound. With additional launch of more robust products, as we mentioned, we have new models as well as refreshed models to be launched. We also have models that are tackling new segments including, for example, extended range energy module. With all that, I think we're very optimistic about our robust growth as well as continued margin improvement. hold the same view as I think communicated to all of you starting two years ago that we'll be breaking even at some point next year, probably towards the end of next year. And that is still a view I hold. I think we hope to deliver on that. And also the improvement also will bring a healthy cash flow for the company as well next year. For example, by the end of this year, we estimate we're going to have over 40 billion RMB on hand. Next year, I think we will still continue to see healthy cash flow, which allows us to have a very comfortable capital base to bring us to break even.

Great. Thanks for sharing that detail and congrats again on the strong results. Thank you.

Thank you. Your next question comes from Ming Shunli with Bank of America. Please go ahead.

Hello, Mr. Xiaopeng and everyone else. Hello, I'm Ming. My first question is about overseas shipping. Taiwan Taiwan My first question is related to export outlook. So in 2024, how do you expect the export sales contribution to your total revenue. And currently, because some overseas markets, the charging infrastructure is not as good as in China. Therefore, do you see any potential bottleneck for EV penetration in certain countries? And in the longer term, will you see your ER EV product to be the major product for the overseas markets?

Hey, Ming. It's Brian again. Yeah, let me address your question on overseas market. First of all, I think we do see overseas market as a very robust growth market for us. It's still very early in the electrification process compared to the Chinese market. And also, I think given our current coverage of covering most of these markets by the end of next year, I think we are hoping to be able to tap into that growth As you mentioned, last year, I mean, I would say this year, our overseas market percentage has increased to around 15% of our sales. I think the next year we expect the contribution will be similar, even though our domestic market growth is very, very significant, but I would still think overseas growth will, I think, has a very similar growth profile as well. And then also in terms of the electric BEV versus extended range format, I think you're right. I think in some markets we do recognize the lack of infrastructure could be a potential bottleneck for BEV penetration. However, I think these markets, I think currently the BEV penetration is still very low. So there is still ample growth opportunity for BEV models themselves. So we are also very hopeful the growth of BEV exports as well as market penetration will increase as we expand into more markets. But at the same time, once we actually have extended range products, we think in some markets, particularly for markets like, let's say, Latin America or Central Asia or Middle East, where charging facility infrastructure is lacking for efficient and fast charging BEV products, some of the extended range products will actually be also attractive. So we're actually very, very optimistic that both BEV as well as extended range products will be finding attractive growth opportunities in various global markets.

Thank you, Brian. My second question is about So my second question is related to capacity. Could you advise your latest capacity and also your effective capacity in 2025? Do you have any plan to expand the new plant or you can just expand your current plant? to meet the demand? And recently, do you also see any component shortage across your supply chain? Thank you.

Hi, Ming. This is Charles. First of all, I think as we mentioned in the earning call that both our Guangzhou and Zhaoxing plant already turned on the second shift. I think each of the plant can support approximately 200,000 to 300,000 per annum based on the two shifts. And also, I think as we communicated before, there are also ample reserved land and also existing plants next to our Guangzhou and Jiaoqing manufacturing base. So we believe that we can expand our production capacity at fast speed and also with low capital intensity. And also, we already had our long-term production capacity planning until 2026. And so, we believe that all these required manufacturing capacity has been well planned ahead. And also, given we have long-term planning for our own manufacturing capacity, we are also working with our suppliers also to expand the suppliers' capacity because as you know that we are pushing really hard on the platform, a unified platform, and also the components sharing across multi-platforms and vehicles. So it is actually more efficient for our suppliers to expand their capacity with us.

Thank you, Charles.

Thank you. Your next question comes from Bin Wang with Deutsche Bank. Please go ahead.

Okay. My first question is about the gross margin of the vehicles. Actually, in the third quarter, you got 2.2 percentage points, much expansion. Can you quantify each of the factors? How much came from the product mix? How much came from cost reduction? How much came from the high base? Because second quarter have some NOB cost. And secondly, okay, you actually died for the number four quarter. Did you think the vehicle gross margin can go to double digit or not? Thank you.

Hey, Ben, this is James. So to your question on the Q3 versus Q2 margin improvement, I'd say it's primarily driven by two aspects. One is we continued engineering cost reduction with regard to efforts on VAVE in combination of the battery cost reduction as we see the battery cost is coming down for the entire industry. You didn't mention the EOP impact. We did have some EOP impact in the second quarter, which is less in the third quarter, which is also driving an improvement of the margin. So that's for the quarter-over-quarter improvement. As we look into Q4, we mentioned earlier the P7 Plus delivery will start in Q4. This is a product that will embed our latest platform with the cost reduction targets achieved. and representing a double-digit gross margin as we communicated earlier. This is going to help us further improve our vehicle margin from Q3 into Q4. So as a trend, we do see margin continue to improve. Combined with larger scale, as Xiaopeng mentioned earlier, we expect our Q4 delivery to be exceeding prior quarters in the history, therefore helping us to thin our manufacturing cost as well and improve overall margin.

Can we have a double-digit cost margin in the number four quarter? Possible?

The overall margin will improve. As you can see, we have an overall margin in Q3 as reported, and you can expect that to improve in the fourth quarter.

Okay, thank you. The question is that recently media reported that Taiwan foundry company may not able to doing the OEM for China chip suppliers actually. for the 7 nanometer. Will this have any potential impact for our upcoming chips? Thank you.

My name is Charles. I think that the mass production of our Turing SOC still progress well. And we haven't seen any impact on our development of the Turing SOC.

Thank you.

Thank you. Your next question comes from Tina Ho with Goldman Sachs. Please go ahead.

如果从我们的ADAS这块支架的智能座舱, 然后支架, Thanks for taking my question. So my first question is regarding long-term cost reduction of EV. So if we look at it from an angle of the powertrain, the ADAS bond, including both smart cabin as well as autonomous driving, as well as for maybe potentially the car body and the interior exterior. So how much potential further cost reduction do you think there is in the longer term? Thank you. 说实话,这个问题从我角度来看也在不断地学习跟成长。

About a year and a half ago, I reported that in this quarter, we hope to achieve a greater cost control. I was still very happy. At that time, I was brave enough to say that. But now we have done it. Then, going back, let's look at the next three years, even four years or so. What else is there? I think from what we see now, there are still a lot of places that are still capable. I think that in addition to the basic standard, what we see in the supply chain, scale, technology, I think there are many types of future models. For example, can a car carry out a larger super layer? Then the super layer can merge many parts of the parts of many parts of the parts. So can you can learn from Apple or other companies to help Tier 1, and even see Tier 2 and Tier 3 together with Tier 1. Their combination of cost, quality, logistics, etc. actually has a lot of space. For example, in terms of electronic materials, what are some of the urgent, comprehensive, I just talked about three or four directions, but from our point of view, it may have seven to eight dimensions in terms of cost control and ability to improve, including the huge ability to improve manufacturing technology. I think these are all companies that will continue to do this in the next three to four years as a company that is innovative in technology. So in the future, Thank you. Thank you for this question. Actually, we've never stopped thinking about that and our understanding and our possible solution to it has been changing over the years.

I remember about a year and a half ago during the earnings call, I actually made a promise of achieving significant cost reduction. At that time, I actually summoned my courage to make that promise, and I'm very happy and proud that we're able to actually deliver what we promised at that time. And in the coming three to four years, obviously, there are a lot of room for improvement when it comes to cost reduction. a lot of things are very obvious for example supply chain optimization scale economy of scale and also technology uh different kind of cost reduction but specifically we can do a lot more things as well for example on the one hand we can do something that we call super integration meaning that we can actually combine different capabilities of different parts together and make something that is significantly different from what we traditionally have or we can also learn and adopt the Apple model, which is to empower the Tier 1 suppliers or help them together develop the capabilities of Tier 2, Tier 3 suppliers' advantages, leveraging their already existing logistic capability and a lot of other details that can help us to improve efficiency and cut costs. In addition to that, we also can look at for example saving and cost control and in the electronic electric materials etc i mean these are just some examples tip of iceberg here really and as a company that's constantly driven by technology innovation we also can look at the upgrade of our manufacturing process our craftsmanship in the coming three to four years we're not going to stop until we achieve you know, the optimal level of cost cutting. Now, in the future AI Tech Day and also in the future earnings call, you can expect to hear our reporting of every year's cost control outcome. And I don't think it's not just going to, I don't think that it will come from only the scale or supply chain control optimization, but more likely being driven by technological innovation. Thank you.

Thank you very much for your answer. And so second question is regarding our 2025 new model pipeline and also volume outlook. Could we get more details in In terms of the four new models, which quarter will they come out, and then what kind of price range, what kind of body type, and also our overall volume outlook for 2025? Thank you.

Hey, Tina. It's Brian. I think, first of all, we are not providing, obviously, any guidance as we've done in the past. So I think right now all I can say is that next year we actually are very confident that we can continue the momentum. We're seeing the second half of this year. And also looking at the growth profile, I think it will be more moderate growth compared to this year. But still I think it will be second half is slightly heavier than the first half. In terms of the model, I think we gave you the total number. We're not at the moment, I think, ready to share specific models and exactly when they will be launched. But we mentioned that there will be four new models. One of those will be extended range model. And also, in addition to those four models, we will have a few refresh of current models. So, and that will be spread over the next four quarters. So, you'll expect to see a new model and refreshment potentially every quarter.

So, for the four new models, our expectation should not be lower versus Mona as well as P7 Plus.

Well, I think we are very, very confident that the models we launch will be leading their respective categories. Obviously, different segments and different categories will have different volume expectations, but we do feel like our models will be very competitive in their respective segments.

Thank you very much.

Thank you. Your next question comes from Nick Lay with JP Morgan. Please go ahead.

Okay, thank you for letting me ask questions. I'm Nick from this department. The first question is about export and shipping. I just mentioned that in the third quarter, the sales of overseas sales accounted for 15% of our third quarter sales, but we see that most of the main factories in overseas prices are usually higher. Its horsepower or single-carburetor is usually the same as the original sales, but it is 1.5 to 2 times. I don't know if this trend is also in line with the situation that Xiao Peng is seeing now. At the same time, Taiwan Taiwan Taiwan Let me translate my question very quickly. At the moment, so-called export overseas market accounted for about 50% of our sales volume, but we understand from other competitors that For those who have overseas exposure, the possibility of profit margin is generally about 1.5 to 2 times higher than the same cost sold in China. It's fair to say the same pattern will apply to S-PoM. And likewise, how do we educate overseas customers that level 2, level 3 functionality is something very nice that they need to have in the future?

Hey, Nick. It's Brian. I think let me address your first question in terms of overseas markets' profitability contribution to us. I think there are a couple areas to think about. One is that, yes, in general, I think the price of selling our models overseas are higher than domestic prices. There's additional costs, obviously, and potential tariff and duties that we have to pay. but the margin in general is slightly higher than the domestic gross margin. But also to be mindful is that the margins that we achieve in a lot of these overseas markets actually are margins that, wholesale margins, because we are working with importers or distributors in those countries where we are not responsible for retailing and distribution of those products. So a lot of those margins actually, it's essentially direct contribution to us rather than just the gross margin. So there's a distinction. But I think going into next year, I think clearly we need to deal with changes in tariff, change in potential new markets, have different regimes. So we need to obviously have a more flexible approach to that structure. But this year, I think the contribution has been pretty positive. And I think Xiaofeng addressed your second question.

Right. The second question, actually, we do see, of course, overseas, there are many different countries. In Europe and China, there is indeed a big difference. For example, users, Chinese users are younger, and European users are more middle-aged. So, for the self-development, or in terms of technology, there is a big difference between China and Europe. But in China and some other countries, there is also a big difference. Let me go back to China and Europe. I believe that from the European point of view, we think that the super-high-speed charging and intelligent service, logistics, and operation, as well as in autonomous vehicles, And the standard LCC and LCC may be the most typical in the sense of automation. But in China's current city-wide automation service market, because it is limited to French rules and many regulations, I believe it will probably be a little bit later than China, even 12 months later. But in fact, we believe that if we do hardware and software ready, In different countries, we will have different combinations of automation. These combinations will make Xiaopeng's high-end brand technology differences stronger. So my core for automation in the future is operation. It will continue to improve and provide stronger automation in a unified standard. And go better around the world. I am very confident. From our data, we are seeing this quickly. Thank you.

Thank you. Yes, indeed. Different countries are different. They have different users, different provinces, different regulations. By comparison, for example, Europe versus China, typically China, as a market, has a younger group of consumers that are really attracted by NDVs, whereas in Europe, the majority of the users are middle-aged and above consumers. But also we see a lot of similarities between China and other parts of the world outside of EU as well. But here for this discussion, I'm going to focus on the differences between Europe and China. Now, right now, when it comes to the capabilities being delivered to our Europe's customers, I think they really love smart cabin, fast and ultra-fast efficient charging, our high-quality services, after-sales services, and also the valet parking, auto parking, LCC, ACC. These are the sort of most daily commonly accepted and preferred ADAS features that are being loved by European customers. And when it comes to other parts of ADAS features, because of the limitation or because of the regulatory environment in Europe, typically the implementation of those features in Europe is about 12 months behind what we can see in China right now. But as a company that is being driven by technology with heavy investment and very strong capabilities in software and hardware, we are very, very confident in adopting the same set of solutions but using different combinations for different market in order to build our global presence as a premium brand going forward. And right now, as has been proven by market feedback, that we are actually able to do that. So in the future, when it comes to landing or implementing our solutions in different countries, We're going to focus more on not only product and service quality, but also the operations across different regions to customize for local consumers. Thank you.

Okay, thank you. My first question is relatively simple. It's about the policy change. Taiwan Taiwan Taiwan My second question is really about the trade policy from top-down standpoint. What's our view on the continuity of the policy into 2025? And likewise, our product sales volume has been very strong. especially for Monarch 3 and PC+. If we place orders today and we only get a card sometime in first quarter next year, what's our marketing strategy for the customer who get a card only in first quarter? Thanks.

Hi, Nick. This is Charles. I think we believe that our newly launched product like MO3 and also the P7 Plus are very competitive in terms of the product capabilities. And I think that the customer chose the product because I think there is no alternative product or very limited options available to them in the price range. So we believe that we will continue to see the strong momentum for the orders for the MO3 and also P7+. I think that more importantly, I think that we have very significant order backlog for both MO3 and the P7+. I think this is also very different from a lot of our peers. We will be carrying probably tens of thousands order backlog for both MO3 and the P7 plus into the Q1. So we believe that that will be a foundation of our growth in 2025.

Charles, my question is, if the client only gets it next year, if the replacement is canceled next year, will the company consider paying the client? My question is for the customers who buy PC Plus, PC Plus or Mono 3, if they only get a car in first quarter of 2025, will Aspon consider reimbursing them, provide additional incentive? Thanks.

Hi, this is Alex. So first of all, we do not have any specific insight about the government subsidies. We expect the auto sector will still be supported by any of the potential stimulus policy as a priority of any of the economic policy. And regarding the customer expectations, I think they have quite reasonable expectations for the delivery time. As you can see, the delivery time for the PCM Plus is 8 to 11 weeks. So I don't think our customers, they are expecting to see sort of delivery before the December of this year. For most of the customers who put their orders right now, they have reasonable expectations. We don't think they will change their decision because of the subsidies. They chose expound cars because of the unique value proposition we bring to these customers in these segments. P7 Plus, more than P03, they exceeded the competitiveness of all our peers' models. So we don't really expect to see a material impact from any potential continuation or discontinuation of subsidies. We just focus on our product competitiveness as well as strengthening our channels.

Thank you.

As there are no further questions now, I'd like to turn the call back over to the company for closing remarks.

Thank you once again for joining us today. If you have further questions, please feel free to contact Xpeng's investor relations through the contact information provided on our website or the Pearson Financial Communications.

This concludes today's conference call. You may now disconnect your line. Thank you.