Li Auto announced its 2026 first-quarter financial results. The company delivered 95,142 vehicles in Q1, representing a 2.5% year-over-year increase. Total revenue reached 23 billion yuan, with cash reserves at the quarter's end standing at 94.3 billion yuan. This marks the tenth consecutive quarter of maintaining cash reserves around the 100-billion-yuan level.
Following the earnings release, Li Auto's management held a conference call. Chairman and CEO Li Xiang, CFO Li Tie, President Ma Donghui, and Chief Technology Officer (CTO) Xie Yan participated and addressed questions from analysts.
The following are the key highlights from the analyst Q&A session:
Morgan Stanley analyst Tim Hsiao: Thank you for taking my questions. I have two. First, regarding our new model, the L9. I just heard Mr. Li Xiang mention that since its launch, Li Auto has received over 10,000 orders. Could you first share how the overall order intake for the L9 is performing? Also, we see the L8 version currently has a delivery cycle of 9-11 weeks. Could you provide an update on the company's current production capacity preparations? Additionally, in the quarterly guidance, what share is attributed to the L9?
Ma Donghui: For the all-new L9, looking at the order structure, the main Livis trim accounts for nearly 90% of orders, while the top-tier Ultra trim makes up about 10%. This fully demonstrates that users recognize the value of advanced technology in the vehicle and are willing to pay for premium configurations. It also signifies that Li Auto's extended-range products have firmly established themselves, achieving a brand breakthrough in the high-end market segment above 500,000 yuan. Moving forward, we will focus on strengthening the product communication for the Ultra trim to further optimize the order mix.
Regarding production capacity, the all-new L9 is produced at our Changzhou manufacturing base, which allows for flexible production alongside our all-new L8, ensuring stable long-term production capacity. May and June are the production ramp-up period, with a monthly capacity of approximately 4,000 to 5,000 units. Currently, the supply of dual-tone bodies and exclusive innovative components for the Livis trim is relatively tight. We have already collaborated with core suppliers to implement supply assurance plans. For the L8 version, production capacity reserves are sufficient, allowing us to flexibly adjust production schedules based on market orders.
Regarding the Q2 delivery volume for the all-new L9, affected by the initial production ramp-up phase, we expect to deliver around 8,000 units. After the production ramp-up is fully completed in Q3, we are confident that the steady-state monthly sales of the all-new L9 will steadily surpass the average monthly level of the previous L9 model.
Morgan Stanley analyst Tim Hsiao: My second question concerns profitability. First, what is the company's expectation for the profit margin in the second quarter? Looking at the full year, where is the turning point? Given the backdrop of rising raw material costs this year, can we return to profitability?
Furthermore, we see that the L6 currently accounts for nearly 60% of the company's sales volume. What is the bottom line for the company's gross margin? Could you also share the gross margin targets for the L9 and other models scheduled for launch later this year?
Li Tie: I'll address this. The Q1 gross margin was affected by multiple factors, including the model refresh cycle, the transition from the L5 to the new version, the increased sales proportion of the L6, and the impact of purchase subsidy policies on the L6. However, with the launch and delivery of the all-new L9, we expect the gross margin to recover to 10% in Q2. From a full-year perspective, with the completion of the refresh for the L-series models and continuous optimization of the product mix, the overall gross margin for this year will steadily improve.
Our current top priority is to successfully complete the refresh of the entire L-series lineup. The all-new L9 fully demonstrates our flagship product strength and technological leadership, receiving strong market recognition and gaining share in the 500,000-yuan price range. The market success of the all-new L9 has established a solid foundation for us in the high-end flagship segment, achieving further breakthroughs beyond the previous L9. This year's all-new L-series product matrix, along with the subsequent all-electric product lineup including the I9, incorporates a significant amount of in-house core technologies, laying a solid foundation for the company's development over the next two years.
CITIC Securities analyst Qiao Wensu: Hello, management. I have two questions. The first is about intelligent technology. Since the deployment of our in-house Maha M100 chip and the Maha VLA large model, what practical effects have been observed so far? What is the feedback? What differences in intelligence does it demonstrate, and what cost-saving effects has it brought? Also, what is the company's next development direction for intelligent driving? This is my first question.
Xie Yan: The Intelligent Driving version 9.0, powered by the in-house developed Maha M100 chip, represents a significant upgrade over version 8.0. The core improvement lies in the vehicle's decision-making capability in complex road conditions. Longitudinal and lateral control is now closer to human driving habits, resulting in a smoother and more comfortable overall driving experience. Version 9.0 is the first intelligent driving version to run entirely on our in-house chip. It is already at the forefront of the highly competitive industry, but this is just the beginning.
Under the new hardware platform, sensors can collect data with higher precision and frame rates. Simultaneously, the powerful computing capability of the Maha M100 supports the operation of larger-scale, higher-quality algorithm models. This new platform achieves synchronous upgrades in data, computing power, and algorithms, driving rapid, leapfrog iterations in our intelligent driving capabilities.
The next steps for intelligent driving development: First, further expand the scale of input data and perception models, allowing more driving-related semantic information to be fed into neural networks, enabling models to identify more dimensional signals from sensors. Second, enhance the model's cognitive ability, focusing on strengthening short-term causal logic learning. This allows the model to move beyond simple behavior replication and make human-like judgment decisions in more complex urban traffic scenarios. Third, optimize the performance of the system's execution end. Through the in-house developed operating system to reduce computing latency, combined with a full wire-controlled chassis, achieve more precise vehicle motion control and faster response speeds, making the intelligent driving system's driving feel more stable and significantly improving safety.
Relying on integrated in-house hardware and software design, our Maha M100 chip achieves a 3x improvement in effective computing power at a cost similar to the previous generation. When compared based on equivalent computing power, the cost is only half that of the previous platform. Under the same model, the input frame rate has doubled, and the inference frame rate has increased even more significantly. Our goal is to achieve performance comparable to Tesla's FSD Beta 14 version in the US by the second half of this year. The high-performance AI inference system built around the Maha M100 lays a solid foundation for us to achieve this target.
CITIC Securities analyst Qiao Wensu: My second question is about the sales end. Since the implementation of our store partner program, what specific changes have been observed in key metrics for the pilot stores—such as store efficiency, average monthly sales per store, per capita output, and expense ratio—compared to before the reform? Has the program's impact on sales met expectations so far? And how does the company quantify the program's expected boost to sales in Q3 and beyond?
Ma Donghui: I'll answer this. Since the launch of the store partner program, by fully delegating operational decision-making authority and profit-sharing rights, we have activated the initiative and operational vitality of frontline teams. Firstly, we can see a significant shift in store managers' operational mindset, transforming from traditional executors into true business operators who can independently assess the input-output of various operations. Focusing on operational efficiency, they conduct refined management, which also enhances the stability of the core management team and their long-term operational commitment. Store partners can deeply cultivate a single store over the long term. Their focus can gradually shift from solely pursuing short-term sales performance to deeply cultivating local user operations, optimizing regional reputation, and building the store's long-term core competitiveness.
Looking at the current operational status, Q1 is traditionally a slow season for the automotive industry, and the partner program is still in its initial implementation phase. On a national average, stores have exceeded their monthly order targets and successfully completed specific repair schedules for an L-series model, also improving user service satisfaction. Moving forward, as store managers accumulate more operational management experience, combined with the company's empowerment training system, the refined operational capabilities of stores nationwide will continue to improve. Thank you.
Goldman Sachs analyst Tina Hou: Thank you very much for your time, management. This is Tina from Goldman Sachs. My first question is about the upcoming Li L8. Is there any information you can share at this point?
Ma Donghui: I'll answer this. Our L-series product matrix will become clearer with the launch of the L8. The L9 focuses on the flagship 6-seater, while the all-new L8 will concentrate on the flagship 5-seater. These two models will form efficient complements, continuously consolidating the brand's advantageous position in the high-end flagship market.
The L8 completed the MIIT new vehicle announcement filing in April. The plan is to officially launch and begin deliveries by the end of June. Compared to the previous model, the new vehicle has increased body dimensions and wheelbase. The interior has been upgraded to a 5-seat layout, significantly optimizing rear passenger space and enhancing the overall riding experience.
Regarding the powertrain, the new vehicle will be equipped with our in-house developed 1.5T extended-range system, paired with a large 72.7 kWh 5C battery. The battery capacity is consistent with the all-new L9, and many technologies are shared from the same platform, resulting in excellent energy consumption and range performance. Additionally, the vehicle offers rich personalized options, including dual-tone bodies and electric side steps. Please stay tuned for our June launch event for more details. Thank you.
Goldman Sachs analyst Tina Hou: My second question is about AI R&D. I'd like to ask management about their views on the current investment landscape and competitive dynamics in the AI industry, as well as the company's own perspective on competition in this field.
Li Xiang: We believe that the competition in the mid-to-high-end intelligent vehicle segment over the next 3-5 years will essentially be a competition in embodied intelligence. The highest technological barrier in the entire industry and the core determinant of a company's long-term competitiveness will be the deep co-design of chips and large models.
Based on our past experience with in-house chip development, technology and information are fluid. When everyone was using Nvidia chips, even though we made many innovations, others could achieve similar levels by poaching our talent.
With our development of more powerful in-house computing chips and the adoption of completely different model deployment methods and model scales, through vertical integration of hardware and software, this method of competition through talent poaching becomes completely ineffective.
Moving forward, we will turn systematic capability into our core barrier. Our capabilities and outputs will no longer be easily taken away by others. Furthermore, time is also a crucial factor. It took us four years from the start of our in-house chip development to its official deployment in vehicles. In the next decade, we will maintain our competitiveness in technology and innovation, and more importantly, maintain technological barriers that are sufficiently far-reaching, sufficiently advanced, sufficiently high, and last for a sufficiently long time.
CICC analyst Jing Chang: Okay, thank you, management, for taking my question. My first question is about embodied intelligence. Mr. Li Xiang just mentioned that we will hold a more detailed, intelligence-focused launch event in June. I'd like to ask if there are any updates regarding the strategy and plans for robotics and embodied intelligence at this stage?
Li Xiang: I think, in the long run, we can clearly see that humanoid robots will be needed in our factories, stores, and by our users. Looking at the market landscape, we believe that startups, mid-sized companies like us with initial revenues around 100 billion yuan, and even larger corporations will all enter the robotics field.
Therefore, robotics does not belong exclusively to us, nor does it belong exclusively to startups. Robots are a standard form of labor. They are products that any company wishing to make a mark in this field will pursue; they are not the exclusive domain of any particular type of company. As long as a company's business requires human labor, it will inevitably manufacture robots. The only difference lies in whether they adopt an outsourcing approach or a self-manufacturing approach.
Furthermore, looking at the final stage, the large-scale commercialization and industrialization of humanoid robots is, for example, somewhat similar to the electric vehicle field between 2010 and 2015. I believe it will take at least three more years, or even longer, to reach that level. So, during this period, there are still many technological challenges to overcome because, in every minor aspect, the future technological path has not yet reached a consensus. These all need to be resolved. That is our assessment.
CICC analyst Jing Chang: The second question is about the overseas market. Could you provide an update on our latest international strategy? Of course, might mid-to-high-end brands face greater challenges when entering overseas markets? Could you also share our plans for 2026 or the longer term, including the pace and progress of overseas expansion, as well as our target key markets?
Ma Donghui: I'll answer this. Li Auto's internationalization strategy is progressing steadily. Regarding overseas markets, we adopt a phased expansion approach. Based on local market size and competitive landscape, we flexibly choose between establishing subsidiaries to develop dealerships or signing exclusive general agency agreements. We aim to rely on leading local partners to quickly establish a complete service system integrating sales, delivery, and after-sales.
Li Auto's brand and products continue to gain recognition in overseas markets. At the Beijing Auto Show in April, we attracted significant attention from overseas media and users. We have officially signed agreements with leading dealers in the UAE and Saudi Arabia. For markets like the Middle East and Central Asia, we will primarily focus on L-series extended-range models. The first overseas-specific version, the all-new L9, will be optimized in hardware and software aspects such as charging performance, cabin air conditioning, and thermal management based on local usage scenarios. It will officially enter the Middle East and Asian markets in Q3.
In May, we will gradually enter markets including Macau (China), Cambodia, Laos, and Myanmar, continuously deepening our presence in Southeast Asia. In the second half of the year, we will also launch the all-electric I6 model in the European market. For right-hand-drive markets, we will introduce the right-hand-drive version of the Mega in core Asia-Pacific regions like Hong Kong (China) and Singapore by the end of the year.
On the product front, we will insist on precise regional adaptation. For all future new models, we will complete adaptation to overseas regulations during the R&D phase to better support Li Auto's internationalization strategy. Thank you. (End)