BYD, the wolf has really arrived

Ask AI · How does BYD’s flash-charging technology reshape the EV refueling and charging experience?

Author｜Eastland

Hero image｜Visual China

On March 25, 2026, BYD (SZ: 002594; HK: 01211) released its 2025 Annual Report. According to the report, 2025 revenue was 804 billion yuan, up 3.5% year over year; attributable net profit was 32.62 billion yuan, down 19% year over year.

In 2025, BYD’s R&D spending reached 63.4 billion yuan. It rolled out a series of breakthroughs, including the “Super e-Platform (maximum charging power 1MW)”, “5th-gen DM (depleted fuel consumption as low as 2.6L)”, “Tianzhen Eye (mass-market intelligent driving, parking as a fallback)”, the “Lingyuan vehicle-mounted drone system”, and the “Cloud Eel-Z intelligent suspension system”……

Whenever BYD releases a piece of deep-tech, onlookers hold their breath to watch market reactions, but the results often end up being a “high opening and a low finish.” On March 5, 2026, BYD released its second-generation Blade battery and flash-charging technology—this time, the wolf really is here.

The underlying logic of “two legs, marching in step”

BYD has long adhered to “two legs, marching in step”: pure electric and plug-in hybrid.

Whether pure-electric driving beats plug-in hybrid, or plug-in hybrid beats pure electric, is the result of multiple factors stacking together—such as oil prices, the progress of charging-station construction, and even fluctuations in the price of battery-grade lithium carbonate (battery costs are inversely proportional to the competitiveness of pure-electric vehicles).

The shallow logic of “two legs, marching in step” is “the east may not shine, but the west will.” This strategy has enabled BYD to consistently become the global sales champion of new-energy vehicles.

Since 2021, the sales curves for pure-electric and plug-in hybrid cars have resembled a “twisted braid”:

In 2021, pure-electric and plug-in hybrid sales were 3.21 million and 2.73 million units, respectively, for a total of 5.94 million units (excluding internal-combustion vehicles); they accounted for 54% and 46% of new-energy passenger car sales, with pure electric leading by 8 percentage points.

In 2022, the situation reversed: pure-electric and plug-in hybrid sales were 9.11 million and 9.46 million units, respectively, totaling 18.57 million units (excluding internal-combustion vehicles); they accounted for 49.1% and 50.9% of new-energy passenger car sales, with plug-in hybrid overtaking pure electric by 1.9 percentage points.

In 2023, the situation reversed again: pure-electric and plug-in hybrid sales were 15.75 million and 14.38 million units, respectively; passenger car sales combined were 30.12 million units (fuel vehicles had been discontinued). They accounted for 52.3% and 47.7% of new-energy passenger car sales, with pure electric leading by 4.5 percentage points.

In 2024, the situation reversed for the third time: plug-in hybrid vehicle sales reached 24.85 million units, up 72.8% year over year, accounting for 58.5% of total sales—leading pure-electric vehicles by 16.9 percentage points.

In 2025, pure-electric and plug-in hybrid sales were 22.56 million and 22.89 million units, respectively. Total passenger car sales were 45.45 million units; pure electric and plug-in hybrid split the market evenly.

In January and February 2026, pure-electric and plug-in hybrid sales were 2.256 million and 2.289 million units, and 1.63 million and 2.3 million units, respectively; plug-in hybrid led by 17.2 percentage points.

In the first two months of 2026, BYD’s sales fell sharply year over year. Among them, pure-electric sales decreased by 35% and plug-in hybrid sales decreased by 36.7%. Has the “two legs marching in step” approach failed? The author believes that the “second-generation Blade battery and flash-charging technology” released on March 5, 2026 has two aspects of impact on sales:

• Before the release

Potential buyers held cash and waited to buy, with a widespread wait-and-see mindset; this was an important reason for the sharp sales drop in January and February 2026.

• After the release

Charging convenience improved, greatly alleviating range anxiety. Consumers who originally planned to buy “plug-in hybrid” may have new options. But there are many countries/regions around the world where it’s not convenient to install charging stations at a large scale, and there are still plenty of older owners who are attached to the in-car experience of fuel vehicles. Over the next few years, plug-in hybrids may not necessarily sell worse than pure-electric vehicles.

A wise approach is to not dwell on it and not be anxious about it—objective conditions + consumer preferences determine, for a particular country/region and during a particular period, which car model is more popular, then you offer that model.

The underlying logic of “two legs, marching in step” is “do good deeds, don’t ask about the future”—focus on refining the technology and solving pain points, and don’t bet on/guess which of pure electric or plug-in hybrid will sell better.

Earning power crushes Tesla

Starting in 2023, BYD’s gross profit amount and gross margin from vehicle sales have exceeded Tesla’s:

In 2023, BYD’s gross profit from vehicle sales surged to 101.6 billion yuan, with a gross margin of 21%; Tesla’s gross profit from vehicle sales fell back to 94.3 billion yuan, with a gross margin of 17.1%; “Bit vs. Bit” flipped, reaching 108%;

In 2024, BYD’s gross profit from vehicle sales continued its growth trend, reaching 137.7 billion yuan and a gross margin of 22.3%; Tesla’s gross profit from vehicle sales further fell to 75.6 billion yuan, with a gross margin of 14.6%; “Bit vs. Bit” reached 182%;

In 2025, BYD and Tesla’s gross profit from vehicle sales were 132.9 billion yuan and 68.2 billion yuan, respectively; BYD was down 3.5% year over year and Tesla was down 10% year over year; BYD’s gross margin was 6 percentage points higher than Tesla’s; “Bit vs. Bit” expanded to 195%;

It’s worth noting that Tesla’s hardware + software revenue is included in vehicle sales revenue—FSD (Full Self-Driving)-related revenue is recorded under vehicle sales. In 2025, FSD revenue recognized from deferred revenue was $956 million, down 19.5% year over year. When Tesla’s vehicle sales are combined with “FSD that earns especially well,” its gross profit is only half of BYD’s gross profit from vehicle sales!

Exporting 1 million vehicles is just the starting point

In 2025, BYD’s biggest highlight was undoubtedly its export business—full-year exports were 1.046 million units, up 1.4 times year over year. Based on publicly available information, the top ten countries by volume were:

Mexico 130,000 units, Brazil 120,000, Belgium 94,000, Indonesia 81,000, the UK 79,000, Australia 48,000, the UAE 47,000, Turkey 46,000, Spain 41,000, and Uzbekistan 28,000. Among these, Germany had 23,000 units, up 700% year over year—being in the top 10 is just a matter of time.

BYD’s overseas selling prices are far higher than those in China; for the same model, the price difference can be several times. Theoretically, exporting 1 million units is equivalent to selling several million units domestically. But in 2025, BYD’s profit did not increase—it actually declined instead.

BYD does not separately disclose sales revenue for domestic versus overseas automotive operations, but BYD Electronics (HK: 00285) reports provide a clue. For example, in 2025:

• Based on:

BYD Group’s overseas revenue was 310.7 billion yuan (including automotive and phone component business);

BYD Electronics’ overseas revenue was 119.0 billion yuan (mainly phone component business);

• Inference:

BYD’s overseas auto revenue was 191.7 billion yuan (317.0 billion - 119.0 billion);

BYD’s domestic auto revenue was 456.9 billion yuan (684.6 billion - 191.7 billion)

• Average selling price:

BYD’s domestic average selling price for automobiles was 128,000 yuan per unit, while the overseas average selling price was 183,000 yuan per unit;

BYD’s push overseas started with commercial vehicles (buses), whose unit price is far higher than passenger vehicles (sedans, SUVs). In the past two years, passenger vehicle exports have surged, pulling down the overseas average selling price.

BYD’s main export models are Yuan PLUS (Atto3), Dolphin, and Song PLUS. The share of premium models such as Han, Fangchengbao, Denza, and Yangwang has been increasing gradually.

If the cost is 96,000 yuan, the selling price is 120,000 yuan, and the export price is 160,000 yuan (an extra 40,000 yuan profit per vehicle). Taxes are 40,000 yuan (about 25%), freight is 15,000 yuan (about 10%), and channel partners are 32,000 yuan (about 20%). Then the overseas terminal price would be close to 250,000 yuan. If profit per car increases by 40,000, wouldn’t exporting 1 million vehicles mean an extra profit of 40 billion yuan?

Actually, it’s not that simple. To open an automobile market in a country, you first have to do market research, right? You need to obtain market access permits, right? You have to set up offices, right…… Building a sales/service network is a prerequisite for selling cars; before a single car is sold, you may already have spent one or two hundred million yuan, and then you still need to run advertising and promotions……

Like automobile manufacturing, automobile sales also have significant scale effects. Take Germany as an example: it has already built 250 sales and service outlets, covering 90% of major cities. Selling 23,000 vehicles in 2025 means it’s impossible to make big money.

BYD’s overseas veteran Liu Xueliang once said: Expanding overseas is not starting from zero—it’s starting from negative numbers.

As of the end of 2025, BYD had entered six continents and 119 countries and regions. Of these, 99% had just crossed the “from negative to zero” threshold. Exporting 1 million vehicles covers the top 10 countries that account for 70% of volume. For the remaining 109 countries, each gets only a few thousand vehicles; selling in such conditions, making money by scalping 180–20 cars should be no problem. But if you follow the official import process and the path of building a sales/service network, reaching break-even at least takes 2–3 years.

One judgment is that exporting 1 million vehicles is only the starting point; if the momentum continues like this, BYD’s overseas business will quickly cross the break-even point. The “money road” is without limit, and high global oil prices will accelerate this process.

Is it “over-researching”

In 2022, BYD’s R&D spending tried to catch up to Tesla. In 2025 it reached 63.4 billion yuan, equivalent to Tesla’s 192%.

From 2014 to 2025, BYD’s total R&D investment over 12 years reached 232 billion yuan.

BYD’s R&D results are abundant: patent applications exceed 71,000, and 42,000 of them have already been authorized.

BYD’s R&D efficiency is not an issue, but if it can’t effectively boost sales, then it amounts to “over-researching.”

For example, take 2025: it held many launch events, from mass-market intelligent driving, to the Qianfu platform and megawatt flash-charging, to hybrid technologies with fuel consumption as low as 2.6L per 100 kilometers, to the vehicle-mounted drone system (Lingyuan). There were also many signature events, such as announcing intelligent parking as a fallback, “blowout-proof” stability for puncture prevention, the fishing hook test results at 210 kilometers, the rollout of the 15 millionth new-energy vehicle, and record-breaking achievements on the Nürburgring Nordschleife track……

Somewhat bizarrely, for all these new technologies released, they didn’t cause much of a ripple; it felt like a mud cow entering the sea. This stands in stark contrast to Xiaomi launch events, where big orders surged like a flood immediately after the event.

People who follow up-and-coming automakers typically number only several million to tens of millions, mainly urban white-collar workers. Once the eagerly awaited technologies are released, they instantly spark heated discussion and high demand within the target group. But at the scale of BYD, Geely, and Chery, the audience for products and technologies extends far beyond the elite group—there are many people, and most are not particularly “moved” by new technologies. They believe in what they can see… once the launch event happens, the big orders “XX hundred thousand” are not something that will definitely occur.

Competitors’ mindset is complex and interesting: they are more “in the know” than the average consumer. Their reaction to BYD’s new technologies is a chain of question marks—“Should we follow?” “Can we follow?” “What if we can’t keep up?”

In the industry, those who can’t sit still throw out “the uselessness theory,” or say “I already knew that.” Those who can sit still patiently observe the market’s reaction…… seeing that BYD’s sales didn’t improve much, those who believe “the technology is useless” claim they’ve won, and then mock BYD for “over-researching.”

In fact, whether it’s new forces or BYD, the key to whether “deep tech” is recognized is whether users can clearly perceive it. For example, the megawatt flash-charging released in March 2025 was tested only on a handful of models such as Tang L, Han L, etc. BYD needs to think through the charging experience mode and communicate it with partners; most consumers have no chance to perceive the convenience of flash-charging.

For another example, at-speed 140 km/h blowout stability: most drivers have never encountered a blowout on a highway in their entire lifetime, so they have no way to perceive it. Besides, most people’s attitude toward safety is that of Lord Ye—paying lip service. Ask them, and they say “we take it very seriously,” “extremely, extremely seriously.” But how many passengers in the back actually buckle their seat belts? Seat belts can indeed save lives, and traffic laws have explicit requirements—but many people still prefer to violate the rules rather than buckle up. If public attitudes toward seat belts are like this, how can blowout stability win against the “big sofa”?

But new technologies are there; sooner or later, the audience will perceive them. And with one “parking as a fallback” on the left and one “ice-and-snow inner-eight” on the right—day after day, small gains accumulating into bigger ones—BYD’s persona of “tech that’s genuinely top-tier” will gradually sink deep into people’s minds.

In short, it’s too early to assert that BYD is over-researching; let’s see what happens to sales over the next one or two years.

The wolf is here

The second-generation Blade battery and flash-charging technology announced on March 5 are installed on nearly all models across BYD’s entire lineup of sub-brands, from below 100,000 yuan to above 1 million yuan.

• Easy to perceive

Charging speed is the easiest to perceive. The launch event doesn’t mention 5C or 10C; it directly promises the charging time—going from 10% battery to 97% takes just 9 minutes, and charging to 70% takes only 5 minutes.

Maybe someone thinks the difference between 9 minutes and 20 minutes isn’t that big. Let’s assume a typical scenario—there are two service windows at a bank branch: window A takes 9 minutes per transaction and window B takes 20 minutes per transaction. Suppose each window receives one customer every 10 minutes. Window A doesn’t require queuing; when a customer arrives, they’re processed immediately, and each customer takes 9 minutes. Window B: the first customer takes 20 minutes; the second takes 30 minutes (including 10 minutes of waiting); the third takes 40 minutes; the fourth takes 50 minutes; the fifth takes 1 hour…… In reality, this situation wouldn’t happen; incoming customers would naturally choose an idle window. But in scenarios where there are both flash-charging stations and non-flash-charging stations, drivers can’t choose like bank customers who want the faster window.

• Changing the business logic of charging

Flash-charging solutions change the investment, operation, and profitability models of charging equipment.

Take Telink Power (SZ: 300001) as an example: it operates 792,000 public charging piles with a market share of 24%. In the first half of 2025, charging volume was 8.5 billion kWh, with service fees of 1.84 billion yuan. Per pile, charging was 62.6 kWh per day, and daily revenue was 13.6 yuan. A 60kW charging pile occupying one parking space theoretically can charge 1,440 kWh per day (60×24), serving 60 pure-electric cars. In reality, the charging amount is only 60 kWh, serving just one vehicle—the efficiency and returns are both quite low. And it wastes a large amount of social resources.

By contrast, for fuel vehicles (taking Beijing as an example): about 1,000 gas stations and 10,000 fuel dispensers serve 5 million fuel vehicles. Suppose each car refuels once every 10 days; each dispenser serves 50 cars per day. The efficiency and returns are 1 to 2 orders of magnitude higher than charging guns. It’s extremely difficult to build a new gas station in Beijing, and acquiring an existing one often costs tens of millions of yuan.

Even if the flash-charging station’s efficiency reaches only half that of a gas station: one minimum unit (two parking spots and two dispensers, plus 400 kWh of storage), serving 40 cars per day, charging 2,400 kWh of electricity, and generating service fee revenue of 720 yuan. At night, it “frees up” 400 kWh of electricity at off-peak valley prices costing 3 jiao (0.3 yuan) per kWh; during the day, it sells at 1 yuan per kWh—earning a spread of 280 yuan. Revenue of 1,000 yuan per 24 hours.

China has fewer than 100,000 gas stations and just over 1 million fuel dispensers, which basically meets the demand of 320 million fuel cars.

As of the end of June 2025, China had 18 million charging piles—18 times the number of gas dispensers—yet it still can’t eliminate range anxiety for 40 million new-energy vehicles. To promote new-energy vehicles, we still need to build charging piles at full scale.

In fact, what restricts charging for new-energy vehicles isn’t that there are too few charging piles, but that charging is too slow. With an average power of only 44kW across the existing 18 million charging piles, even without queueing, charging 60 kWh still takes 1.5 hours.

One flash-charging pile is equivalent to 10 regular piles, which can greatly reduce the number of newly built charging piles while significantly improving the economic efficiency of building and operating charging piles.

• A 10GWh energy storage network

Flash-charging stations are equipped with energy-storage facilities with extremely large capacity and ultra-high charge/discharge power, distributed across the country—there’s limitless room for imagination. By the end of 2026, 20,000 flash-charging stations will be completed, with total storage capacity of about 10GkWh (among 18,000 stations, the “in-station” stations have 400 kWh each, and 2,000 standalone stations have 1,000 kWh each).

In the past, every time BYD released a new technology, the auto industry would stage a “the wolf is coming” scenario. Now, the wolf really is here!

*The analysis above is for reference only and does not constitute any investment advice!

End