Li Auto executives respond to seven key questions about self-developed batteries

Author | Zhou Zhiyu

Li Auto's recent shift from CATL to self-developed batteries has attracted market attention.

With the i9, MEGA, and other models successively announcing battery transition plans, some users still prefer CATL batteries. Whether Li Auto's self-developed batteries can achieve the same level of performance and quality, the reasons for this transition, and how an automaker can compete with a professional battery manufacturer have become key questions from the outside world.

On September 21, Ma Donghui, co-founder, president and chief engineer of Li Auto, and other executives gave a concentrated response to these issues in an exchange with media outlets such as Wall Street Insights.

"It's not a backup plan." Ma Donghui made it clear that starting with the new generation Li Auto L8, Li Auto is pushing forward the use of self-developed batteries in new products, and this strategic direction has been determined.

Liu Liguo, Senior Vice President of Electric Vehicle R&D at Li Auto and Chairman of Shandong Li Auto Battery Co., Ltd., stated that Li Auto's self-developed battery products, manufactured by Sunwoda, have after-sales quality performance that is no weaker than that of any other battery supplier; however, Li Auto still lags behind CATL in R&D capabilities for cutting-edge technologies such as solid-state batteries.

From the necessity, technical level, and cost of self-developed batteries, to users' trust in battery brands, and the boundaries of cooperation with suppliers, Ideal's executive system answers seven key questions.

I. Why develop in-house ?

"This strategic direction is very clear; it's not a backup plan," Ma Donghui said.

According to his introduction, Li Auto established its "dual-energy strategy" of intelligence and electric power as early as 2020, with the electric power direction including two platforms: range-extended electric vehicles and pure electric vehicles. The launch of self-developed and self-manufactured products this year is a phased result after six years of continuous investment.

Why expand the application of self-developed batteries at this time? Ma Donghui explained that the technical conditions and self-made PACK production lines are already in place, and the company's current investment scale can support this strategy.

Li Auto's self-developed technology has a clear division of labor in production: the battery cells are developed under Li Auto's leadership and manufactured by partners; the PACK, or battery pack, is developed and manufactured by Li Auto itself. The manufacturing of battery cells still requires the manufacturing and process capabilities of partners.

Ma Donghui summarized the necessity of self-developed technology as "three connections," corresponding to data, vehicle systems, and service systems. Under the traditional procurement model, automakers cannot obtain complete information on cell formulas, process parameters, and testing data for each cell; what they receive is more like a "gray box." Li Auto hopes to connect manufacturing data with vehicle and user usage data to identify problems before installation and delivery, and to continuously monitor battery health and lifespan trends after installation.

The application of 5C batteries highlights the importance of system-wide coordination in vehicles. The compatibility of the supercharging network, and the operation of the electric drive, thermal management, and charging systems, all affect the battery's actual performance in the vehicle. Ma Donghui believes these aspects need to be designed holistically.

The integration of the service system connects product definition, service standards, and after-sales processing. When batteries experience issues such as range degradation, automakers, battery manufacturers, and users may use different standards, and the problems may involve cells, PACKs, integration, or software. Li Auto hopes to unify after-sales standards, allowing users to directly contact automakers to resolve issues.

"Users' reasonable demand is that when quality standards are not met, they don't need to care which link the problem lies in; Ideal should take full responsibility," said Ma Donghui.

II. What is the actual level of our self-developed batteries ?

When comparing Li Auto's self-developed products with CATL's, Liu Liguo first mentioned 420,000 vehicles. This represents the number of vehicles in use after Li Auto's self-developed batteries, manufactured by Sunwoda, entered the market with models like the Li L8 and L7. He stated that the overall after-sales quality performance of this batch of products is no weaker than that of any other battery supplier.

"So, in terms of the current products, I don't think there's any difference," Liu Liguo said. As for the self-developed and self-manufactured batteries, they will be officially used in the new L8 model this year.

When discussing cutting-edge technologies such as solid-state batteries, he admitted that there is indeed a significant gap between Li Auto and CATL. CATL focuses on the entire industry, while Li Auto invests in its own product experience; the scope and scale of their R&D efforts are different.

Liu Liguo also used this to explain the difference in the size of the two companies' R&D teams. Professional battery companies serve multiple vehicle customers and also cover energy storage, different material systems, and charging rates; Li Auto is currently focusing its product development resources on 5C supercharging, with teams specializing in vehicle materials and thermal management also participating.

The widely circulated claim that Li Auto has invested approximately 500 million yuan in battery R&D over the past five years has been clarified during the discussion. Liu Zhimin, Senior Director of Power Batteries at Li Auto, stated that the 500 million yuan only covers expenses related to laboratory material input, testing, and verification. Li Auto's annual battery R&D expenditure exceeds 200 million yuan, with a cumulative investment of over one billion yuan over the past six years, in addition to investments in fixed assets.

According to Liu Zhimin, the research and development began with prototype verification of positive and negative electrode materials, electrolytes, and formulation systems. Once the technology matured, it was then transformed into engineering solutions with partners, with both ternary lithium and lithium iron phosphate routes being pursued simultaneously. On the manufacturing side, the cells and PACKs have a total of over 9,400 control standards, supplemented by online testing and AI anomaly identification. He stated that the current defect rate in cell manufacturing has reached PPB, or one part per billion.

The power performance under low-temperature conditions is an example cited by Liu Liguo. In the past, the entire vehicle was calibrated according to the discharge power table provided by the battery manufacturer; after self-development, the state of charge and available power can be calculated more accurately, and with the control algorithms of the whole vehicle and the battery, the problem of insufficient power under low temperature conditions can be alleviated.

Liu Zhimin also introduced the battery pack's heat dissipation design: a dual-sided cooling scheme using both the terminal surfaces and the bottom surface. During research and development, it was discovered that directly cooling the terminal surfaces can more efficiently remove heat.

III. Why directly anchor at 15C ?

Li Auto envisions the next stage of supercharging technology as advancing the capability of "recharging 500 kilometers of range in 10 minutes" to "recharging 500 kilometers of range in 5 minutes." The 15C specification is a technical requirement proposed based on this performance goal.

Liu Liguo explained that when choosing the 5C model, Li Auto considered battery capacity, vehicle heat dissipation, charging station construction, and power supply conditions at charging stations. Taking a 100 kWh battery as an example, at 5C, the peak charging power would exceed 500 kW.

According to him, most of Li Auto's more than 4,000 charging stations have a power output of around 600 kilowatts, with many others at 480 kilowatts or even lower; less than 5% of the stations can provide a power output of around 800 kilowatts. Even if the battery has a higher rate of charge, the station may not be able to provide enough power.

Vehicle-side heat dissipation is also a constraint. Ma Donghui gave an example of a high-temperature travel scenario: After a long journey, the battery temperature is already high. After entering a service area, passengers remain in the vehicle, the outside temperature reaches 40 degrees Celsius, the air conditioning continues to cool, and the battery begins high-power charging. At this time, multiple systems are vying for heat dissipation capacity.

The charging pile also requires a continuous output of high current. Whether relying on direct grid power or adding energy storage, higher efficiency rates require more infrastructure investment. Ma Donghui believes that if the goal is simply to reduce the charging time from about 10 minutes to 8 minutes or 6 minutes, the cost of the new products and charging stations must be calculated to determine if it is worthwhile. Ideal hopes that the next investment will bring the charging efficiency closer to that of refueling a gasoline vehicle, and therefore is focusing its R&D resources on 15C.

"Our goal is not to achieve the 15C rating, but to achieve 500 kilometers of range with just 5 minutes of charging," said Liu Liguo.

Currently, 15C is still in the technology development stage. Ma Donghui stated that the timing of its application will depend on the maturity of the technology, infrastructure, and user value, and it has not yet entered into specific product planning. Liu Liguo predicts that 5C will still have high user value in the next two to five years.

Li Auto is also continuing to improve the existing vehicle-charging pile integration. According to Liu Liguo, the self-developed batteries and self-built charging piles can exchange more charging detection data beyond the standard charging protocol. The charging piles have already deployed Orin SoC (System-on-a-Chip), which, with stronger computing power, improves the charging experience and reliability of its own vehicles at its own charging piles.

IV. Do users really care about battery brands? How to deal with " de-branding " ?

Ma Donghui believes that consumers' attention to battery brands ultimately corresponds to the safety, lifespan, performance, and service guarantees behind the brand.

However, he also acknowledged that some users prefer CATL batteries. "This is also the user's right to choose," he said, adding that not all consumers understand the industry situation and the ideal self-developed process.

According to Ma Donghui, Li Auto had already issued an announcement during the launch of the i9 and MEGA models explaining the battery switchover arrangements and clearly informing each user of the battery brand corresponding to their order. If this does not align with user preferences, the company respects the user's choice.

Liu Liguo also admitted that the new force has been established for a short time, and the self-development of core components faces challenges in brand recognition. The experience of participating in product definition and technology development in the past needs time for consumers to understand.

Air springs went through a similar process. Ma Donghui recalled that when they first wanted to develop L8 and L9, they ideally chose to cooperate with domestic companies such as Konghui Technology and Baolong. At that time, these suppliers had limited brand awareness in the relevant fields, which later increased as the products were applied.

"The so-called 'de-branding' essentially means the rise of new brands," he said. Batteries also need to build trust through charging speed, durability, range, winter degradation, and after-sales service. Ultimately, Li Auto is responsible for quality, and this trust "needs time to prove itself."

V. Where does the cost advantage of self-developed batteries come from? How can costs be reduced without economies of scale ?

In response to questions about its smaller scale compared to leading battery companies, Ma Donghui broke down costs into five parts: production line amortization, manufacturing costs, raw materials, logistics and warehousing, and R&D investment.

He emphasized that Ideal will not reduce costs by replacing cheaper materials, but rather hopes to improve efficiency in all aspects through product platformization and collaborative R&D and manufacturing, while ensuring performance and quality.

Liu Zhimin proposed a solution: standardize battery cell specifications by using one or two types of lithium iron phosphate cells and one or two types of ternary lithium cells to cover all vehicle models, and then adjusting the number of cells connected in series to match different power requirements. For different vehicle models requiring the same amount of power, the battery pack should be shared as much as possible.

Ma Donghui explained that the platformization of battery cells allows different car models to reuse production lines as much as possible, improving utilization and reducing depreciation costs allocated to individual products.

During the manufacturing process, Li Auto relies on its intelligent industrial team and the Li Auto Lianshan system to monitor and analyze production data, identify waste, and improve the pass rate. According to Liu Zhimin, AI inspection has been deployed on the production line, and the AI algorithms are integrated with the Lianshan system to identify abnormal trends. A unified battery system can also reduce redundant configurations in warehousing, packaging, tooling, and transportation.

Regarding raw materials, Ma Donghui stated that in the past, they were more passive in the face of price fluctuations; after developing and manufacturing their own materials and outsourcing, Ideal hopes to participate more proactively in upstream materials and risk management, and improve their ability to manage cost fluctuations and supply stability.

Self-developed technology does not mean saving on R&D costs. According to Ma Donghui, this cost used to be borne by suppliers and reflected in the purchase price; now, it is borne by Li Auto, with partners focusing on manufacturing and processes. Whether it's contract manufacturing or outsourcing, a reasonable profit margin should be reserved for partners.

Production capacity will follow the demand from vehicle manufacturers. Ma Donghui stated that the plan is linked to the launch date and sales targets for new products in 2027, with the PACK production line being built according to vehicle demand. Any surplus cells produced for Ideal in the battery cell production line can be used by Sunwoda for other brands.

Regarding whether to utilize a second brand to absorb production capacity, he stated that Li Auto has no plans for a second brand.

6. What is the strategic positioning of self-developed batteries? Will there be open cooperation in the future ?

"Our goal is to refine the battery capabilities to a level that fully supports the user experience of our products, so that we can define and be responsible for the battery performance of each car ourselves," said Liu Liguo.

Specifically regarding the collaboration, Li Auto hopes to first achieve PACK platformization on the vehicle platform before making customized requests for battery cells. According to Liu Liguo, regardless of who the supplier is, development must be carried out according to Li Auto's system and size standards to meet the requirements of data, experience, and service synergy.

Ma Donghui summarized the criteria for selecting suppliers as TQCD, namely technology, quality, cost, and supply and delivery capabilities. He also mentioned that the willingness of partners to collaborate deeply in R&D, manufacturing, and data is becoming increasingly important.

Sunwoda and Zhongchuang Innovation Aviation have both joined this cooperation system. According to Ma Donghui, both companies adopt the model of Li Auto developing its own battery cells and having their partners manufacture them, with no difference in quality standards. Li Auto's cooperation with Sunwoda started earlier, and they have an investment relationship and have also established a PACK joint venture; the products developed in cooperation with Zhongchuang Innovation Aviation have been included in the Ministry of Industry and Information Technology's announcement and will be used in the new Li Auto i6 in 2026.

As for why he didn't pursue the same self-developed manufacturing cooperation with CATL, Ma Donghui didn't elaborate, only reiterating that excellent suppliers meeting the criteria are welcome to cooperate. When asked whether battery manufacturers would differentiate their pricing based on whether automakers have self-developed capabilities, he stated that he hadn't heard of such a thing and had no experience with it.

He cited the example of electric drives: even after Li Auto shifted from purchasing complete electric drive systems to developing and manufacturing them in-house, it would still purchase rotors and other components from Inovance. After developing its own systems, the scope of cooperation could be adjusted.

Ma Donghui also discussed applications beyond automobiles. He views batteries, electric drives, and drive-by-wire chassis as the underlying technologies for embodied intelligent products to operate in the physical world. Cars, robots, and smart glasses all require energy systems, and some cutting-edge technologies not yet suitable for power batteries can be explored in other products.

7. What does the trend of OEMs developing their own batteries mean ?

Liu Liguo used the analogy of a gasoline-powered car engine. Early on, domestic brands used third-party engines, but as their products developed, they gradually built their own R&D and manufacturing capabilities. From the perspective of core capabilities and product definition, he believes that power batteries may also undergo a similar process.

In his assessment, as battery technology matures, the gaps in basic performance, reliability, and quality between different products are narrowing, making it increasingly difficult to differentiate products based on a single battery component. Automakers need to consider the overall user experience that the battery delivers when working in conjunction with other vehicle systems.

Ma Donghui further discussed the vehicle architecture. He explained that the low-voltage battery, DC-DC converter, power distribution unit, and thermal management systems—which functions are suitable for integration into the battery and which should be separated—require a unified design across the entire vehicle. This will affect how components are connected and which functions are assigned to which components.

He used the analogy of a fully steerable chassis: steering, braking, and suspension involve lateral, longitudinal, and vertical control respectively, and are traditionally supplied by different vendors, but some complete driving experiences rely on the coordination of all three. A similar relationship exists between the battery and the various electrical systems of the vehicle.

According to the path disclosed by Li Auto, while automakers are increasingly involved in battery R&D, the need for specialized battery companies has not been eliminated. Cell manufacturing and processes are still provided by partners, while Li Auto is more closely integrating material solutions, PACK design, control software, and vehicle development.

This allows the collaboration between the two parties to move further into the front end of product development: the battery cells need to be adapted to the vehicle platform, the manufacturing data needs to be integrated into vehicle usage and quality management, and the production line also needs to be arranged according to the vehicle model requirements.

Li Auto hopes to change, through its self-developed technology, how these processes collaborate and who ultimately delivers the complete vehicle experience to the user.

Risk Warning and DisclaimerInvesting involves risk; please exercise caution. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article are suitable for their specific circumstances. Any investment decisions made based on this information are at your own risk.