different paths lead to the same destination.

arts｜"chinese entrepreneurs" reporter wang wentong

edit｜ma jiying

header image source｜visual china

as nio (9866.hk) handed in its first-half report card, the performance of nio, xiaobian and li auto in the first half of 2024 came to light.

wei, xiao and li, with different financial performances, all mentioned two things in their financial reports and recent press conferences:one is the release of new cars and price reductions; the other is the increase in smart driving.

in the first half of the year, both xpeng motors and nio launched new brands or product lines, exploring the price range of products. for example, nio launched its second brand ledao, which focuses on family cars. at the semi-annual earnings conference, li bin announced that the price of ledao's first model l60 will be lower than the pre-sale price of 219,900 yuan, and the final price will be announced on september 19, and will be delivered on a large scale at the end of the year; xpeng motors launched mona m03, with the lowest price dropping to less than 120,000 yuan.

after the failure of mega, ideal auto also launched low-priced models and cut prices. the ideal l6, which was launched in the first half of the year, isideal autothe first model priced below 300,000 yuan; followed by price cuts for ideal's entire 2024 product line, with the l series dropping by 18,000 to 20,000 yuan and ideal mega dropping by 30,000 yuan.

entering july,wei, xiao and li auto all frequently speak out on intelligent driving: ideal auto’s summer smart driving conference on july 5,nioat the nio in 2024 press conference and xpeng's ai intelligent driving technology press conference on july 30, "opening the city" and "end-to-end" became the terms they talked about the most.

ideal auto's per capita revenue is 1.7 times that of xpeng motors and 2.2 times that of nio.

in the first half of 2024, the per capita revenue of ideal auto (revenue/number of employees) is 1.8604 million yuan, which is about 1.7 times that of xiaopeng motors and about 2.2 times that of weilai in the same period. in the first half of 2023, the per capita revenue of ideal auto is about 4 times that of xiaopeng motors and about 3 times that of weilai.

in terms of revenue, all three companies achieved growth.nio's revenue in the first half of the year was 27.355 billion yuan, a year-on-year increase of 40.65%; xpeng motors' revenue in the first half of the year was 14.660 billion yuan, a year-on-year increase of 61.16%; li auto's revenue in the first half of the year was 57.312 billion yuan, a year-on-year increase of 20.81%.

in terms of profits, only ideal auto recorded a positive net profit in the first half of the year, but it also declined; the other two companies recorded negative values, but they all narrowed. ideal auto's net profit in the first half of the year was 1.695 billion yuan, a year-on-year decrease of 47.40%; xpeng's net profit in the first half of the year was -2.653 billion yuan, a year-on-year decrease of 48.41%; weilai's net profit in the first half of the year was -10.231 billion yuan, a year-on-year decrease of 5.22%.

in terms of gross profit margin, the gross profit margins of xpeng motors and nio's automotive divisions both increased in the second quarter, but were still lower than those of tesla and li auto; although li auto's divisional gross profit margin declined, it has exceeded tesla for two consecutive years.

after a fierce price war in the first half of the year, weili's per-vehicle cost (vehicle sales cost/sales volume), per-vehicle revenue (vehicle sales revenue/sales volume) and per-vehicle gross profit (vehicle sales gross profit/sales volume) have also changed.

the cost of each bicycle of wei, xiaopeng and li auto is similar, with the average cost of each bicycle being about 230,000 yuan.

ideal auto's per-vehicle revenue and per-vehicle gross profit both declined, but were still higher than those of xpeng auto and nio in the same period, at 288,900 yuan and 54,000 yuan respectively; xpeng auto's per-vehicle revenue and per-vehicle gross profit both increased, with its per-vehicle gross profit turning from loss to profit, but were still the lowest among the three companies in the same period, at 238,000 yuan and 15,200 yuan respectively; nio's per-vehicle profit increased but revenue did not, with per-vehicle revenue down 25,800 yuan year-on-year and per-vehicle gross profit up 8,000 yuan year-on-year.

ideal auto can achieve higher efficiency in nio, xiaobian and li auto, firstly because its sales volume is larger and its business is more focused; secondly because it has less financial and debt pressure.

in terms of capital and debt, ideal auto has more cash reserves (including cash and cash equivalents, restricted cash, short-term investments and time deposits) and a higher current ratio (current assets/current liabilities, representing short-term debt repayment ability, generally best at around 1.5~2).

however, behind the achievements, ideal auto also has hidden worries.judging from the inventory situation, in the first half of 2024, ideal auto's inventory was 8.308 billion yuan, about 1.5 times that of xpeng and about 1.7 times that of nio in the same period.

at the same time, ideal auto's profits this quarter were supported by financial management. following the operating loss in the first quarter, ideal continued to maintain an operating loss in the second quarter, with a loss of 117 million yuan. however, thanks to ideal auto's abundant cash and performance in investment and financial management, ideal auto received more than 1.4 billion yuan in interest income andinvestment incometherefore, the overall profit was nearly 1.69 billion yuan, a year-on-year decrease of 47.8%.

the same tough battle

for nio, xiaopeng and li auto, all three have a tough battle to fight in the field of intelligent driving.

zhang xiang, a senior automotive industry analyst, believes that intelligent driving has now become an important selling point for car sales. in order to attract consumers and not want to be eliminated by the wave of intelligent driving, car companies are investing more and more in intelligent driving.

mei songlin, a senior analyst in the automotive industry, believes that if the development of new energy vehicles is compared to a pyramid, the base is the three electrics (battery, motor, and electronic control), the middle is the cost performance, and the top of the pyramid is the level of intelligent driving. now we have entered the era of intelligent driving competition.

he xiaopenghe also stated publicly that in the last decade, car-buying decisions were influenced by a series of factors such as car size, power, and safety, but in the next decade, the weight of ai in car-buying decisions will be close to 50%.

to some extent, the competition in intelligent driving is a battle of funds.

in its semi-annual report, ideal auto stated that self-developed core technologies are an important cornerstone for the company's long-term development, and that it will invest 3 billion yuan in technology research and development for two consecutive quarters in 2024. lang xianpeng, vice president of intelligent driving research and development at ideal auto, previously mentioned that the training computing power required to ultimately achieve autonomous driving must reach the level of 100 eflops, which translates to an annual investment of more than $1 billion.

xpeng motors stated that it has invested 3.5 billion yuan in ai training and will invest more than 700 million yuan each year in computing power training in the future.

at the earnings conference, nio emphasized that competition in smart driving is still fierce, and in each quarter, nio still maintains an investment intensity of approximately 3 billion yuan on a non-gaap basis.

the financial report shows that in the first half of 2024, nio's r&d investment was 6.083 billion yuan, 2.817 billion yuan, and 6.076 billion yuan, respectively, a year-on-year decrease of -4.59%, 5.78%, and 41.86%.

behind the heavy investment, a resource competition is beginning, and the competition includes data and computing power.

according to public information, nio, xiaopeng and li auto have currently accumulated computing power resources as follows: at the 2024 chengdu auto show, ideal auto announced that its training computing power had reached 5.39eflops, and it is expected to exceed 8eflops by the end of 2024; on august 4, nio announced that the total computing power of its smart driving end cloud was 306.9eops; at the ai ​​smart driving technology launch conference on july 30, xpeng motors announced that it had an ai computing power reserve of 2.51eflops.

in terms of data, zhang xiang pointed out that the larger the number of cars a car company owns, the more data it collects and the faster the training speed. different models of vehicles from the same car company may all use the same intelligent driving system, and the collected data can be shared.

lang xianpeng also stated at the 2024 chengdu auto show that in the past two months, the sales of ideal auto models equipped with the ad max intelligent driving system have maintained an increase of more than 10% every month. among models priced above 300,000 yuan, the sales of models equipped with the ad max intelligent driving system accounted for 70%, and in some models and some regions, the proportion of models equipped with the ad max intelligent driving system reached more than 90%.

he also said that the outside world criticized the "matryoshka" appearance of the ideal l series, but this is precisely the advantage of the ideal l series in intelligent driving. the sensor layout and sensor model of this series are exactly the same, and the data can be fully reused, which is very beneficial to the research and development of autonomous driving.

xpeng motors and nio are also stepping up the release of new cars. xpeng motors mentioned in its semi-annual report that starting with the launch of mona m03 in august, the company is about to enter a strong product cycle, and in the next three years, a large number of new and modified models will be launched.

li binat the 2024 semi-annual financial report meeting, it was stated that the second factory producing ledao and weilai will achieve double-shift production at the end of september and the beginning of october. the third factory will officially start production in the third quarter of next year. this year, ledao plans to deliver 20,000 units.

an important pressure for nio, xiaopeng and li auto to accelerate their smart driving layout comes from tesla.

in may 2023, musk announced that tesla would release an end-to-end large-scale autonomous driving system. ten months later, in march 2024, tesla launched fsd v12. on april 6, musk announced that by the end of 2024, tesla would invest more than $10 billion in autonomous driving for training computing, huge data pipelines, and massive video storage.

according to cailianshe, tesla's artificial intelligence team announced on a social platform on september 5 that it expects to launch a fully autonomous driving (fsd) system in china and europe in the first quarter of 2025, but it still needs regulatory approval.