On November 1, each company reported October deliveries: Leapmotor 106,127 units, standing above 100,000 for the fourth consecutive month; XPeng 50,237 units; Xiaomi over 50,000, both new highs for the year; NIO, Harmony Intelligent Mobility Alliance, and Li Auto all below 35,000. The first place is more than double the second place.

Four years ago it was not like this. In 2021 Leapmotor delivered only 43,700 units for the whole year, ranking last among the major new forces. Many people still did not pay much attention to this company, but it had already built up both volume and reputation.

Leapmotor's ability to take first place lies in keeping prices down while the cars still sell well and the reputation is not bad: high-level smart driving does not require an extra payment, a large number of core components are made in-house, and costs are kept under control. Brand awareness is not the main reason.

1. October: First Place Was Built on Volume

Figure 1 October 2026 deliveries of new forces (announced by each company on November 1; Xiaomi only disclosed "over 50,000")

Leapmotor's October sales rose 51% year on year, with cumulative sales of 773,000 units from January to October, including exports. XPeng rose 20% year on year; NIO's three brands combined totaled 34,246 units, down 15.2%; Harmony Intelligent Mobility Alliance 33,537 units, down 50.8%; Li Auto 29,025 units, down 8.6%.

In July Leapmotor delivered 101,267 units, becoming the first new force to break 100,000 in a single month, and it has not fallen below that since. The saying "Leapmotor has passed 100,000 while no one else has passed 50,000" was no longer accurate by October: XPeng and Xiaomi had just crossed the 50,000 line. But the two combined are only about equal to Leapmotor alone, and the first place is 3.7 times the last place.

Where the volume comes from depends on the volume-driving end.

2. For Just Over 80,000 Yuan, LiDAR Comes Standard

Figure 2 Leapmotor A10 in the metal. Guide price 65,800–86,800 yuan, with LiDAR on the top trim

Leapmotor has four series, A, B, C, and D, covering 60,000 to 300,000 yuan, and the volume driver is the A series. The A10 launched in March this year has a guide price of 65,800–86,800 yuan; the top trim comes with a Hesai LiDAR and Qualcomm 8650 (about 200 TOPS), enabling parking-space-to-parking-space navigation assist. In October the A series sold nearly 40,000 units, accounting for nearly 40% of the month, and the A10's average retail price was about 76,000 yuan.

Previously, if you wanted high-level smart driving under 100,000 yuan, you either had to pay extra for options or switch to a more expensive car. Leapmotor eliminated that step: within the same budget, both equipment and smart driving are provided. Zhou Hongtao, head of the electronics product line, said that at the end of 2023 the company judged that smart driving had become a must-have for car buyers, and the A10 is the result of that judgment applied to a volume model.

Overseas sales are also adding volume. In 2023 Stellantis invested about 1.5 billion euros for a stake of about 21%, and in the Leapmotor International joint venture Stellantis holds 51%. In the first half, exports were 96,000 units, 27% of total sales; in October the B10 went into production at the Zaragoza plant in Spain. In the first half, vehicle and component revenue was 35.6 billion yuan on 356,000 deliveries, averaging about 100,000 yuan per vehicle—it built solid orders through scale and equipment, not by raising unit prices.

3. Gross Margin Is Not High, but Money Is Spent Frugally

Figure 3 Display unit of Leapmotor's LEAP 3.0 architecture (2023 Guangzhou Auto Show): central computing platform, zone controllers, and battery pack (Source: Tim Wu / Wikimedia Commons, CC BY-SA 4.0)

In 2025 Leapmotor's revenue was 64.73 billion yuan, gross margin rose from 8.4% the previous year to 14.5%, and net profit was 540 million yuan, making it the second new force to be profitable for a full year according to its announcement; fourth-quarter gross margin was 15.0%, a single-quarter high. In the first half of this year revenue was 38.11 billion yuan, gross margin fell to 11.7%, and net profit was 210 million yuan, which the company attributed to rising raw material prices and a higher share of the cheaper A series; in the second quarter it returned to 12.6%. In August, full-year net profit guidance was cut from about 5 billion to about 3 billion yuan, and vehicle gross margin guidance was 10%–11%.

Compared with peers, Leapmotor's gross margin is actually on the low side. XPeng's second-quarter overall gross margin was 20.7%, yet it still lost 1.34 billion yuan; NIO's was 18.4%, yet it still lost 528 million yuan and only barely turned positive after adjustments; Li Auto's vehicle gross margin was 9.4%, with a loss of 1.705 billion yuan. The difference is in expenses: Leapmotor's sales, administrative, and R&D expenses combined were 5.32 billion yuan in the first half, about 14% of revenue; XPeng spent 2.91 billion yuan on R&D in a single quarter, a 27.4% expense ratio; NIO's second-quarter expenses were about 20% of revenue. The gross margin is thinner, but expenses are more frugal, prices are set low, and it can still make money—others who want to match this price may not be able to match its cost structure.

Keeping expenses down relies on in-house development and manufacturing. According to Zhou Hongtao, more than 60% of components are developed and produced in-house, all controllers are developed in-house, the 18 core component factories basically invest only in equipment and do not build plants, and investment in the three electric systems is about 3.6 billion yuan. Of purchased headlights, 20%–30% is the supplier's gross margin, which can be saved by making them in-house. In 2025 compressors and thermal management also entered mass production, and seats and bumpers are being converted to in-house production. What should be bought is still bought: chips use Qualcomm, and in 2020, the toughest time, the self-developed Lingxin 01 was discontinued; LiDAR uses Hesai, the chassis is not developed in-house, and a test track alone would cost 500–600 million yuan. Internal project approval requires payback within three years.

But the quality of profit still needs to be seen clearly. In the first half, operating profit was only 130 million yuan, while "other income" was 1.08 billion yuan, and carbon credit income also grew very fast with exports. Only when vehicle gross margin stops declining will profitability be considered solid.

4. A Dahua Background Is the Confidence Behind Its Willingness to Do Things Itself

Figure 4