Hello, ladies and gentlemen. Thank you for standing by for NIO Inc.'s Fourth Quarter and Full Year 2025 Earnings Conference Call. Today's conference call is being recorded. I will now turn the call over to your host, Mr. Rui Chen, Head of Investor Relations and Corporate Finance of the company. Please go ahead, Rui.

Good morning and good evening, everyone. Welcome to NIO's Fourth Quarter and Full Year 2025 Earnings Conference Call. The company's financial and operating results were published in the press release earlier today and are posted on the company's IR website. On today's call, we have Mr. William Li, Founder, Chairman of the Board and the Chief Executive Officer; and Mr. Stanley Qu, Chief Financial Officer. Before we continue, please be kindly reminded that today's discussion will contain forward-looking statements made under the safe harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. Forward-looking statements involve inherent risks and uncertainties. As such, the company's actual results may be materially different from the views expressed today. Further information regarding risks and uncertainties is included in certain filings of the company with the U.S. Securities and Exchange Commission, the Stock Exchange of Hong Kong Limited and the Singapore Exchange Securities Trading Limited. The company does not assume any obligation to update any forward-looking statements, except as required under applicable law. Please also note that NIO's earnings press release and this conference call include discussions of unaudited GAAP financial information as well as unaudited non-GAAP financial measures. Please refer to NIO's press release, which contains a reconciliation of the unaudited non-GAAP measures to comparable GAAP measures. With that, I will now turn the call over to our CEO, Mr. William Li. William, please go ahead.

Hello, everyone. Thank you for joining NIO Inc.'s 2025 Q4 and full year earnings call. In Q4 2025, we delivered 124,807 smart EVs, achieving a year-over-year increase of 71.7%. Quarterly deliveries for all three brands reached a record high. For the full year 2025, the NIO, ONVO and FIREFLY brands worked in close synergy and delivered a total of 326,028 vehicles, up 46.9% year-over-year. This marks our return to a strong growth trajectory. In January and February of 2026, we delivered 27,182 and 20,797 vehicles, respectively. For the first quarter, we expected total deliveries to be between 80,000 and 83,000 vehicles, representing a year-over-year increase of 90.1% to 97.2%. Turning to the financial performance. In Q4 2025, our vehicle margin reached 18.1%. The continuous improvement in margin was mainly driven by strong sales growth, a higher mix of high-margin models and continued vehicle cost optimization. Meanwhile, the margin of other sales reached 11.9%, supported by the expanding scale and improving profitability of our services and community-related business as our user base continues to grow. In Q4, non-GAAP operating profit reached RMB 1.25 billion, while GAAP operating profit reached RMB 810 million. On the cash flow side, we delivered positive free cash flow for two consecutive quarters and achieved positive operating cash flow for the full year of 2025. This quarterly profitability also further validates the competitiveness of our technology roadmap, products and business models. It also demonstrates the continued strengthening of our operational capabilities and efficiency, providing a solid foundation for long-term sustainable development. Since 2025, we've been serving a broader group of users through the NIO, ONVO and FIREFLY brands. The three brands have been well recognized by users in their respective segments. For the NIO brand, the All-New ES8 has shown strong market momentum since deliveries began at the end of September 2025. In just 160 days, it reached the milestone of 70,000 deliveries, setting a monthly delivery record among vehicles priced above RMB 400,000. In Q2 this year, we will introduce the NIO ES9, an executive flagship SUV. As a combination of NIO's 11 years of technological innovation and user experience development, the NIO ES9 is equipped with multiple industry-leading technologies, delivering a luxurious executive mobility experience for users. In addition, the ET5, ET5T, ES6 and EC6 will introduce the 2026 versions in Q2. Overall, the NIO brand will continue strengthening the product lineup and its position in the premium segment. For the ONVO brand, its L90 has been very well received by large-family users since its launch. Thanks to its innovative experience, outstanding safety performance and exceptional space design, the ONVO L90 became the best-selling large battery electric SUV in 2025. Next, the ONVO L80, a large 5-seat SUV with front and rear independent suspension, will be introduced in Q2. At the same time, L90 and L60 will also receive product upgrades and refreshes. With an expanding product lineup and continuous product upgrades, the ONVO brand will deliver the finest mobility experiences to more family users, driving steady sales growth. The FIREFLY brand officially launched in 2025 has quickly established itself, thanks to its clear market positioning, cost-leading product competitiveness and innovative user co-creation model. In 2025, the FIREFLY ranked number one in the high-end small car market for seven consecutive months. User interest in the FIREFLY continues to grow across more regions. Our long-term investments in core technologies are beginning to bear fruit. Key technologies, such as the world's first automotive-grade 5-nanometer chip for Smart Driving, the full-domain vehicle operating system and the SkyRide Intelligent Chassis have all achieved mass production. These innovations not only elevate product performance and user experience, but also deliver significant cost advantages. Together, they form a strong foundation for the company's long-term competitiveness. In late January this year, we released a new version of NIO World Model known as NWM. In this version, we adopted a training paradigm with complete closed-loop reinforcement learning, one of the first implementations of its kind in China. With the advanced architecture of the world model and its closed-loop reinforcement learning, NWM has significantly enhanced the performance of Navigate on Pilot, or NOP, across all domains. February marked the first full month following the rollout of the new version. During the month, the share of driving time using Smart Driving increased by more than 80% compared with January. With this update, over 2,000 power swap stations in urban areas now support power swap pilot, seamlessly integrating with urban NOP and offering users the industry's only fully automated recharging experience. ONVO Smart Driving has also achieved end-to-end upgrades across Navigate on Autopilot for urban roads, parking assist and Smart Safety. This year, ONVO will further upgrade the Smart Driving hardware and software systems across its lineup, allowing more users to benefit from our full-stack proprietary technologies. On the sales side, we currently operate 171 NIO Houses and 395 NIO Spaces, while ONVO has 420 stores. For the service network, we have 406 service centers and 75 delivery centers. In 2026, the three brands will continue to strengthen their presence in key markets while expanding into more lower-tier markets through Sky stores, which serve as shared sales and service stores for all three brands. As of now, we have 3,815 power swap stations and more than 28,000 power chargers and destination chargers worldwide. On February 6, 2026, the company reached the milestone of 100 million cumulative swaps, which demonstrates that battery swapping has been widely validated by both users and the market. During the Chinese New Year holiday, our swap volume reached new highs, surpassing 177,000 swaps in a single day. Power Swap is a systematic and innovative solution to addressing the mismatched life cycles of vehicles and batteries; serving also as a form of energy storage, it is increasingly becoming an integral part of the new power system. Over time, the continued expansion of the power swap network will enhance the EV user experience and provide a unique defensible competitive advantage for the company's long-term success. While continuing to deepen our presence in the Chinese market, we are also steadily advancing our global strategy. So far, the FIREFLY, including its right-hand drive version, is available in ten countries. In 2026, we will continue to expand overseas through the country distributor model with FIREFLY serving as a leading brand. Last month, our Smart Driving chip subsidiary, GeniTech or Shenji, signed an agreement with its first round of equity financing. The round raised RMB 2.257 billion, giving Shenji a post-money valuation of more than RMB 8 billion. Shenji's chip technology, R&D capabilities and commercialization outlook have been widely recognized by multiple leading industry investors and top market institutions. This financing will support Shenji's continued development of high-performance chip products and strengthen our long-term positioning in autonomous driving and embodied AI. The company has now entered the third stage of its development, embarking on a new cycle of high-speed growth. We will continue to make determined and sustained investments in the 12 full-stack technology domains for smart EVs, ensuring our leadership in both products and technology. In 2026, we will launch three new models, further strengthening our product portfolio and expanding our share in the premium large vehicle segment. We will also continue investing in charging and swapping infrastructure, while improving the commercial operations of our infrastructure network. In addition, we will further strengthen our sales and service network, ensuring that our overall system remains resilient in an increasingly competitive market. In 2026, we will continue to deepen organizational transformation, refine the CBU mechanism centered on user value creation and strengthen our company-wide business system with heightened ROI awareness and improved cost control. We remain committed to our original aspiration as a user enterprise, delivering more competitive products and technology, better services and stronger business performance to honor the trust and support of our users, shareholders and investors.

Thank you, William. Let's now review our key financial results for the fourth quarter of 2025. Our total revenues reached RMB 34.7 billion, up 75.9% year-over-year and 59% quarter-over-quarter. Vehicle sales were RMB 31.6 billion, representing a strong increase of 80.9% year-over-year and 64.6% quarter-over-quarter. The year-over-year and quarter-over-quarter growth were mainly due to increased deliveries and higher average selling price driven by positive product mix effect. Other sales were RMB 3 billion, up 36.6% year-over-year and 17.5% quarter-over-quarter. The year-over-year growth was driven by increased sales of used cars, technical R&D services and sales of parts, accessories and after-sales vehicle services while the quarter-over-quarter increase was mainly due to the increase in revenues from technical R&D services and sales of parts, accessories and after-sales vehicle services. Looking at margins. Vehicle margin was 18.1% compared with 13.1% in Q4 last year and 14.7% in the last quarter. The year-over-year and quarter-over-quarter improvement were driven by positive product mix effect with the ramp-up of higher-margin products. Moreover, gross margin from other sales reached a record high of 11.9% during the quarter, reflecting the continued enhancements in the profitability of our user base-driven service and community-related businesses. With the improvement in both vehicle margin and other sales margins, overall gross margin increased to 17.5% compared with 11.7% in Q4 last year, and 13.9% in the last quarter. With the implementation of sales business unit mechanism and the progress of organizational optimization, we are seeing a sustained improvement in operational efficiency. R&D expenses were RMB 2 billion, decreased 44.3% year-over-year and 15.3% quarter-over-quarter. The year-over-year decrease was mainly driven by lower personnel costs in R&D functions due to organizational optimization and decreased design and development costs from different development stages. The quarter-over-quarter decrease was primarily driven by lower personnel costs in R&D functions due to organizational optimization. SG&A expenses were RMB 3.5 billion, decreased 27.5% year-over-year and 15.5% quarter-over-quarter. The year-over-year and quarter-over-quarter decreases were mainly driven by a decrease in personnel and related costs in marketing and other supporting functions as a result of organizational optimization as well as the decrease in sales and marketing activities. In the fourth quarter, we achieved an important milestone, our first ever quarterly profit. Profit from operations was RMB 0.8 billion compared with loss from operations of RMB 6 billion in Q4 last year and RMB 3.5 billion last quarter. Excluding share-based compensation expenses, adjusted profit from operations was RMB 1.25 billion. Net profit was RMB 0.3 billion compared with net loss of RMB 7.1 billion in Q4 last year and RMB 3.5 billion last quarter. Excluding share-based payment compensation expenses, adjusted net profit was RMB 0.7 billion. Furthermore, we delivered positive operating cash flow and positive free cash flow this quarter. We ended this quarter with a stronger balance sheet with total cash and cash equivalents, restricted cash, short-term investments and long-term time deposits amounting to RMB 45.9 billion. That wraps up our prepared remarks. For more information and the details of our unaudited fourth quarter and full year 2025 financial results, please refer to our earnings press release. Now I will turn the call over to the operator to start our Q&A session. Operator?

Your first question comes from Tim Hsiao with Morgan Stanley.

This is Tim from Morgan Stanley. Congratulations for achieving the first ever quarterly profitability. I have two questions. The first question is about the product and the volume sales growth target. I guess, during the call, I think William just mentioned he is going to launch ES5 series and the 6 series facelift and L80 in the second quarter. So what's our plan for second half? Could we have a quick update on the model pipeline for second half? And separately, given the challenging auto industry backdrop, is NIO still maintaining annual volume growth target of 40% to 50% this year? That's my first question. My second question is about autonomous driving because I think during the opening remarks, William mentioned after the rollout of the world model, the share of AD usage grew 80% month-over-month in February. So what key feedback has been received regarding user experience? Additionally, we noticed several EV peers have or would introduce more meaningful system model upgrades this year. What would be NIO's key technological highlights and differentiators, especially in the tight Level 3, Level 4 range? That's my second question.

Thank you for the question. It's true that the entire industry has encountered challenges in the first quarter of this year as a lot of the industry numbers and data have already shown the market environment. For this year, we also believe that the size of the Chinese passenger vehicle market will see a slight decline from last year. But within China's passenger vehicle market, we believe that the penetration rate of new energy vehicles will continue to grow. And among the new energy vehicles, battery electric vehicles will see stronger growth momentum. Last year, the increase of the new energy vehicle market as a whole was largely driven by BEV models. The BEV penetration rate increased from 26% to 33%. As technology and infrastructure networks continue to mature, we believe that the BEV market segment will maintain strong momentum this year. More specifically for the BEV segment, we also see good potential within the premium segment as perception and awareness of premium BEV models have already been established. For vehicles priced above RMB 300,000, we saw an increase in BEV penetration of 58% year-over-year, while for REEV models, penetration in that segment decreased by 4%. Overall BEV penetration in the premium segment increased from 14% in Q4 2024 to 27% in Q4 2025. More specifically, large three-row BEV SUVs and large five-seater BEV SUVs also saw strong momentum last year. Since September last year, the large three-row BEV SUV segment led across all powertrains for five consecutive months. In the second half of last year, BEV sales increased by 350%, while REEV decreased by 6%. We believe that with this trend where BEV is growing within the segment, our product launch cadence is well matched with the macro trend in the industry. This supports our confidence for continued growth this year. As mentioned, in Q2 on April 9, we will host a technology launch event for the NIO ES9, our flagship executive SUV and technological flagship; the official launch will be in Q2. In Q3, we will introduce a large five-seater SUV developed on the same platform as the All-New ES8. In Q2 this year, ONVO will introduce the L80, a large five-seater SUV. Together with the L90 and the All-New ES8, which are already in the market and popular among users, we will have five midsize and mid-to-large SUV models to complete our product lineup in the large premium SUV sector. This will be a key driver for our full-year growth. Also, ET5, ET5T, ES6 and EC6 are leading in their respective segments with steady demand. ONVO L60 is top three among peers in its segment and will receive upgrades soon. FIREFLY is performing well in the high-end small car segment with significant market share, and since the Spring Festival holiday we have seen rebounding demand. For the full year, we are confident in achieving year-over-year volume growth of 40% to 50%. The first two months of the year are normally a seasonal low for the industry, but even against this backdrop, we achieved year-over-year growth in our sales volume. For Q1 guidance, we target over 90% year-over-year growth. Regarding autonomous driving, we believe there are two key parameters to showcase if Smart Driving experience and functionality are best for our users. The first is the share of Smart Driving time among the total driving time in real-life scenarios. The second is the number of accidents mitigated or avoided by Smart Driving. From January to February, after the world model rollout, Smart Driving usage increased by more than 80% month-over-month. This is based on the new world model version with limited computing investments made last year. This year, we will increase investments into computing power resources and roll out more training sessions for the new world model. We are targeting two major releases in Q2 and Q4 this year. Overall, the new architecture, featuring the world model and reinforcement learning, is capable of delivering a good experience and has greater potential for continuous capability and functionality upgrades. A large amount of data is also helping with experience improvement. We will have these two major releases.

Your next question comes from Bin Wang with Deutsche Bank.

I have two questions. Number one is about the first quarter margin guidance. Based on your volume assumption, what should we look at? Because you're actually facing rising memory cost and maybe some of the raw material costs. Recently, we may actually see some subsidy changes for ESA, about RMB 10,000. So can you provide a first quarter margin guidance? And the second question is more about receivables from related parties because investors see this number keep increasing every single quarter. Can you provide a forecast what's the trend for these receivables from related parties and exactly where they came from, such as your battery swap partners?

Thank you for the question. Overall, for the vehicle gross margin in Q1, it will be maintained at a similar level as in Q4 last year. In Q1, there are seasonality factors as well as policy impacts. But overall, we still have ES8 order backlogs, and since the Spring Festival holidays, we see a good recovery in ES8 order momentum. The ES8 model will still make up a major part of our total deliveries in Q1. With a product mix largely driven by high-margin products like the ES8, we will be able to maintain a relatively flat vehicle gross margin in Q1. As you mentioned, there are external factors, including rising raw material costs, rising memory chip costs and lithium carbonate costs for battery packs. Those impacts will start to take effect in Q1. But since they have just begun, the current impact on our Q1 vehicle margin is limited. Regarding your question on accounts receivable from related parties, it's mainly the receivables by our battery asset management company, NIO Power. In recent months, we are seeing growth in both sales volume of our products as well as the take rate of Battery as a Service. We also see increasing receivables by the battery asset management company. They have been keeping diversified and smooth financing channels through debt financing, equity financing and bank-backed consumer installment financing. Overall, we believe we will have good control over the receivables by the battery asset management company and the payment terms.

Your next question comes from Paul Gong with UBS.

My first question is continuing with raw material cost inflation. Stanley mentioned that the pressure is just beginning to be felt in Q1 and probably in Q2 more pressure could be felt. Given NIO is positioned relatively high end and still has relatively strong demand and new orders compared to peers, do you think you can pass this part of the cost inflation to downstream customers? That is my first question regarding pass-through of raw material cost inflation. My second question is regarding expense. We have seen that cost control has taken effect, especially R&D which has reduced from RMB 3 billion in Q2 to RMB 2 billion in Q4. Given Q4 is traditionally the strong season for expenses, moving into Q1 and Q2, do you foresee further optimization on the expense side? Or do you think RMB 2 billion per quarter is the new low?

I will follow up on my previous answer. For this year, we do face pressures in the vehicle cost structure. Because of the trend toward AI and increased computing power needs, as well as geopolitical tensions, we see rising material costs and volatility in prices for memory chips, copper, lithium carbonate and other raw materials. So we do have pressure on cost structure and vehicle margin. For the full year, we do not yet have a clear picture regarding the full implications of all these volatilities and uncertainties. In the meantime, we will keep working with our supply chain partners to improve supply chain efficiency and mitigate negative impacts. As introduced, we will launch three new large models this year. Together, we will have five large SUV models in our lineup, and they will make major contributions to sales and delivery volume since larger vehicles normally have higher margins and are more resilient to raw material volatility. We are confident we can mitigate or offset the impact of rising raw material costs and try to maintain our vehicle gross margin within a reasonable range for the full year. Regarding R&D, in 2026, on a quarterly basis, we will maintain R&D investment at around RMB 2 billion to RMB 2.5 billion per quarter. So for the full year, R&D expenses will be at a similar level to 2025. With the cell business unit mechanism in place, we will use that to improve R&D efficiency, eliminate low-efficiency items, and improve ROI. We will dynamically manage R&D expenses based on business performance and ROI analysis so that we don't compromise investments into key products and technologies and our long-term competitiveness.

Your next question comes from Nick Lai with JPMorgan.

I'd like to follow up on previous questions on profit and cost. Given William talked about a very strong volume outlook for 2026 with many new higher-priced, higher-margin products from second quarter onwards, but at the same time cost inflation pressure will kick in from second quarter as well. Putting this together, can you help us understand whether we should expect NIO to potentially reach non-GAAP profit or breakeven sometime in the second half? And what does that mean for free cash flow position for the full year?

Thank you for the question. Regarding sales volume, we remain on track for the target of achieving 40% to 50% year-over-year growth in sales volume. As mentioned, we will launch three large models this year, which gives us confidence in the volume target. Regarding vehicle gross margin, we do have pressure on margin due to cost structure. But as mentioned, we will introduce more large models this year that will contribute significant margin. For example, ES8 in Q4 last year had vehicle gross margin close to 25%. Overall, large models are more resilient to price volatility. With the volume and margin outlook combined, we still target achieving full-year non-GAAP operating profit breakeven in 2026.

My second question is really about the Shenji self-designed chip. After recent successful fundraising of RMB 2.257 billion, can you share with us the chip strategy in the medium term? Aside from supporting internal sales, is it fair to say Shenji will also try to explore external customers like other OEMs or other suppliers?

For Shenji, it has completed its first round of equity financing and received wide recognition by professional and industry institutions, which speak highly of Shenji's capabilities and R&D. Ongoing R&D activities will develop our next-generation high-performance chips for our products. In the meantime, Shenji is planning and exploring mid-end chips that are more approachable for more clients in the industry, including potential partners and clients in robotaxi and embodied AI applications, where we see a good fit. We are already exploring possibilities with some external clients, including some automotive companies that have shown interest in our chip products. The second chip product from Shenji, which is a high-end but more cost-competitive processing chip suitable for a broader client base, has already achieved a successful tape-out and is preparing for mass production. It is a very competitive chip product.

Your next question comes from Jing Chang with CICC.

My first question is about the upcoming NIO model, the ES9. Compared with ET9 released last year, could you elaborate on its new capability enhancements? And also how is the pricing position for this new vehicle? My second question is about the energy replenishment method. There has been more market discussion recently. Against the rapid development of fast charging, could you elaborate on the advantages and necessity of battery swapping, especially regarding infrastructure construction and your mid- and long-term charging and swapping construction plan?

Thank you for the question. ET9 is our executive flagship sedan and the most premium and advanced product in our lineup. Recently, we released the Horizon Editions for the ET9 and they were well received. The ES9 carries over advanced technologies and features from the ET9, but it also includes unique tech innovations. Positioning-wise, these two products are different: ET9 is an executive sedan and ES9 is a technology flagship as well as a flagship SUV model. Regarding energy replenishment, we are happy to see more automotive companies participating in charging and swapping network construction. More players joining joint infrastructure development can accelerate BEV adoption and the conversion from ICE and REEV to BEV. Charging and swapping are not contradictory. We have developed more than 28,000 power chargers and destination chargers and are one of the most active Chinese car companies in charging network development. Home charging, supercharging and swapping each serve different purposes for different use cases. We offer a holistic solution including charging, swapping and upgradability. While peers announce faster charging technologies, in the foreseeable future, charging will not outperform the speed or experience of power swapping. Swapping provides additional benefits that charging cannot. Swapping is a systematic solution to solve mismatched life cycles between vehicles and batteries. In China, battery warranties are generally eight years or 160,000 kilometers or at 70% state of health. Vehicle useful life for private users can be up to 15 years, so battery life is shorter than vehicle life. If batteries and vehicles are tightly coupled, the mismatched lifecycle becomes a problem for users. Power swap enables a flexible solution to this mismatch. Under our power swap system combined with long-life batteries, charging and discharging strategies, life cycle monitoring and battery health inspections, we maintain battery safety and maximize longevity. This requires systematic thinking and design. Beyond speed, power swap is also a distributed energy storage system. With renewable energy and green electricity generation, storage and consumption are challenges. The power swap station is part of a NIO power system that helps energy storage, consumption and interaction with the power grid. Swap stations can reduce secondary energy loss and improve efficiency by around 6% compared to supercharging that requires repeated charging. Economically, swapping is efficient and viable. If our installed 3,800 power swap stations each had a full complement of battery packs, the total energy storage would be around 6 to 7 GWh, which is sizable. Overall, power swap not only provides a better recharging experience for users but also helps our long-term business development and is highly defensive.

Your next question comes from Ming-Hsun Lee with Bank of America.

I have two follow-ups. My first question is related to the chip business. You mentioned the second chip will be more powerful and competitive than the first. Can you give guidance on this chip? In the future, will you only focus on autonomous driving and humanoid robot chips, or will you also develop smart cabin chips? Also, because you start to adopt your chip, you save costs. Could you give guidance about vehicle model coverage for your chip and how much percent cost can be saved? My second question is related to your service business. In Q4, your service revenue gross margin reached 11.9%. Is that mainly because utilization of your swap stations continued to increase? Looking into 2026, do you expect the gross margin of the service business will continue to improve based on your projection?

Thank you for the question. The second chip product from Shenji will also be a 5-nanometer automotive-grade chip, but it comes with a more competitive and efficient cost structure. Performance-wise, it will be on par with leading industry chips. Cost-wise, it will be more approachable and efficient than our first-generation Shenji 9031 chip. We are focusing on on-device inference chips rather than training chips. The chip can be used for autonomous driving, smart driving and embodied AI applications. In terms of which vehicle models will be equipped with our second chip, we cannot disclose specific model-level details, especially if it relates to external clients, but we have industry peers and other clients showing strong interest. Regarding service revenue, other sales are mainly service revenues and community-related revenues. In 2025, as our user base and operational efficiency grew, other sales revenue exceeded RMB 10 billion and became a growing part of total revenue. This year, as our user base continues to grow, we expect other sales revenue to grow and make a positive contribution to group-level gross margin. Regarding charging and swapping, while we continue to deploy infrastructure—for example, installing around 1,000 new power swap stations per year—near-term upfront investment will mean operational losses on the charging and swapping network side. However, profit from other sales can essentially cover those losses. The overall development will be conducted at a controlled pace. In 2025, we achieved profitability in other sales, mainly services and community-related businesses, and in 2026 we believe we will continue to enhance profitability in this segment even while installing 1,000 new power swap stations.

Your next question comes from Yuqian Ding with HSBC.

I have two questions. First, to follow up on OpEx optimization. We talked about R&D guidance before. Could you share more color in terms of SG&A? In Q4, we hit SG&A absolute value at RMB 3.5 billion. Can we extrapolate to understand that's the baseline, or a range we can use for 2026? Second, could you share volume and margin breakdown by brand? We talked about volume targets and reasonable margin. Could you help us understand between the three portfolio brands or between the core five big SUV portfolio versus the rest? Give us a sense of dynamics between brands and how we cluster volume mix to support sustainable margin expansion.

Thank you for the questions. In Q4 2025, as we saw growth in sales volume and revenue, SG&A as a percentage of revenue fell into a reasonable range. This was driven by our CBU mechanism and organizational efficiency improvements. Also in Q4, we did not launch new models, so we did not have significant marketing expenses. For 2026, we will continue CBU efforts to ensure SG&A expense efficiency and strong returns. The absolute amount of SG&A will grow, but proportionally to sales revenue increase, we aim to control SG&A to be within 10% of sales revenue. Regarding volume and margin breakdown, the market will be dynamic in 2026, so it's difficult now to give precise breakdowns by model. High-level, the five large SUV models will play a major role in sales volume. FIREFLY has rebounded since the Spring Festival and remains a volume driver. For vehicle gross margin targets over a longer time horizon, we aim for 20% to 25% vehicle gross margin for the NIO brand, above 15% for the ONVO brand and above 10% for the FIREFLY brand. Additionally, our management mindset has shifted from purely volume-driven to finding a balance between sales volume, vehicle gross margin and overall business performance, driven by our CBU mechanism. Each vehicle model now has a dedicated vehicle strategy team responsible for business performance, balancing margin and volume. This was a major change since implementing CBU and is reflected in our improved performance from Q4 to Q1.

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

Thank you again for joining us today. If you have further questions, please feel free to contact NIO's IR team through the contact information on the website. This concludes the conference call. You may now disconnect your lines. Thank you.