Automakers building their own energy replenishment stations is not a new phenomenon.
From Tesla's first Supercharger stations in China in 2014 to NIO's first battery swap station completed in 2018, energy replenishment has always kept pace with electric vehicle sales.
However, since 2026, the pace has clearly accelerated.
Data from the China Passenger Car Association (CPCA) shows that the retail penetration rate of NEVs reached 67% in September, hitting a historic high. The energy replenishment network is shifting from being a "sales accessory" to becoming the main battlefield of industry competition.
In September this year, Geely Holding and NIO announced cross-shareholding. Subsequently, in October, Yinwang Intelligent Technology, CATL, and Times Electric Service signed the Strategic Cooperation Agreement on Intelligent Energy Replenishment Ecosystem in Ningde.
At the same time, BYD's flash-charging stations have exceeded 12,165, while Tesla's Supercharger stations in China have surpassed 2,500.
From the capital binding between Geely and NIO to BYD's ten-thousand-station scale, the way automakers deploy energy replenishment networks is undergoing significant changes.
Yang Xi, an industry veteran who has worked in the energy replenishment sector for 13 years, told Gasgoo: "In the early days, automakers built self-owned replenishment networks primarily to 'avoid the awkwardness of selling cars.' Now, energy replenishment is becoming a key lever for differentiated competition among automakers."
From "Sales Accessory" to "Strategic Positioning": Three Leaps in Energy Replenishment Logic
The most direct reason automakers initially built charging networks was the insufficiency of public charging infrastructure. Without them, electric vehicles simply wouldn't sell.
Around 2014, when Tesla entered China, its first batch of Supercharger stations followed suit. In 2018, NIO completed its first battery swap station, incorporating "swappable and chargeable" into user rights.
During this stage, energy replenishment served as basic infrastructure supporting vehicle sales.
According to Yang Xi, the goal of automakers' self-built networks at that time was to "avoid the awkwardness of selling cars." Energy replenishment was treated as a cost item, not a business opportunity.
The second shift occurred around 2025.
Ultra-fast charging technology matured rapidly: BYD's second-generation Blade Battery reduced charging time to under 9 minutes and launched the "Flash Charging China" initiative. Reportedly, BYD completed its 10,000th flash-charging station in August 2026, reaching 12,165 by October, with a year-end target of 20,000, plus another 6,000 overseas. Tesla's Supercharger stations in China exceeded 2,500, opening over 950 to non-Tesla owners.
During this stage, energy replenishment began to play a role in enhancing owner stickiness and feeding back into vehicle sales.
Reportedly, among BYD's flash-charging users, nearly one-third are owners of other brands.
Yang Xi analyzes that this indicates relatively weak brand-binding effects for plug-in charging: owners do not refuse to use stations due to mismatches between the station brand and their vehicle brand. However, this does not mean charging networks lack value: Tesla's Supercharger system has long been viewed as part of its brand moat, and BYD's ten-thousand-station flash-charging network is also building user stickiness.
The third shift occurred in 2026.
Geely and NIO bound their battery-swap assets through equity, targeting 10,000 swap stations by 2030 with annual power demand exceeding 10 billion kWh; Yinwang, together with CATL and Times Electric Service, integrated Qiankun intelligent driving into the ChocolateSwap network; CATL equipped all its swap stations with ultra-fast chargers, attempting to define both battery-swap and ultra-fast charging standards simultaneously.
For the first time, energy replenishment became a strategic asset contested by automakers and supply chain giants.
Behind this transformation lies a repeatedly validated figure: anxiety about energy replenishment accounts for approximately one-third of the weight in purchase decisions for users switching from ICE to EVs. Domestic electrification penetration rates above 80% are directly determined by the coverage and service capacity of energy replenishment infrastructure.
Therefore, the energy replenishment network itself is a function of sales volume. This is why automakers are willing to invest billions of CNY in businesses that are not profitable in the short term.
Tesla's investment trajectory confirms this. Since entering China in 2013, Tesla has never ceased investing in energy replenishment: as of March 2026, it had more than 2,500 Supercharger stations and over 12,000 Superchargers in China; this year, it launched 55 highway service area Supercharger stations and 212 V4 Superchargers in Chongqing at once.
From "sales accessory" to "differentiated service" and then to "strategic positioning," the positioning of automakers' energy replenishment efforts is upgrading.
So, what exactly are automakers paying for?
Why Are Automakers Flocking to Energy Replenishment?
According to Yang Xi, there are four main motivations for automakers deploying energy replenishment networks.
First, brand tonality. Automakers want to convey the brand perception of being an "energy service provider" to users. Energy replenishment networks are among the few assets capable of carrying dual imaginations of "technology" and "service." Tesla's Supercharger stations and NIO's battery-swap system essentially serve brand narratives.
Second, alleviating purchase anxiety. Anxiety about energy replenishment for users switching from ICE to EVs accounts for about one-third of purchase decision weight, and domestic electrification penetration rates above 80% are determined by the coverage and service capacity of energy replenishment infrastructure. Reportedly, BYD launched "Dream Fulfillment Stations": if enough owners apply within designated areas, the brand builds a station locally. By the milestone of 10,000 stations, behind the 4,228 Dream Fulfillment Stations were 74,281 user applications for construction.
Third, supporting capital market narratives. For listed automakers, energy replenishment provides ready-made material for a second growth curve. When whole-vehicle sales face pressure, they can still talk about energy services, battery assets, and electricity trading. This narrative line directly supports capital expectations regarding automakers' "business diversity."
Fourth, finding new user touchpoints. Whole-vehicle sales and standardized services are approaching saturation, leaving energy replenishment as one of the few scenarios where new connections can be established. A detail in the Geely-NIO collaboration illustrates this: NIO President Qin Lihong estimated that relying on the battery-swap network, a commercial vehicle could save up to 90 minutes per day, translating to a 15%-20% increase in operational income.
In terms of path selection, automakers currently mainly adopt three models: BYD, Tesla, and Li Auto choose self-building, independently funding and retaining service dominance; Xiaomi collaborates with third-party charging pile enterprises to offer exclusive charging privileges without bearing heavy assets; XPeng adopts a hybrid model of limited self-building combined with third-party partnerships, balancing cost and coverage.
A new change in 2026 is that these paths are beginning to "merge." For example, the cross-shareholding between Geely and NIO essentially uses equity to integrate two networks and two standards into one.
The boundaries between the three paths are loosening. Tesla, which has the heaviest self-build approach, has opened nearly 40% of its Supercharger stations to the industry and offers non-Tesla owners benefits such as time-of-use pricing and parking fee reductions; BYD co-builds the flash-charging ecosystem with Sinopec, PetroChina, Shell, TELD, etc., sharing heavy assets with partners while retaining technology and traffic entry points.
But behind path selection, the barrier logic of the two replenishment methods differs.
Reportedly, battery-swap technology possesses strong exclusivity: battery pack standards, swap structures, and user ecosystems are all built upon the same technical system. The certainty that "once you enter, you will definitely get a share" is the core reason why NIO, CATL, and Yinwang continue to double down on battery swapping.
The barriers of charging networks lie on another level: Tesla's Supercharger system has long been viewed as a brand moat, and BYD's ten-thousand-station flash charging is also binding user stickiness. The difference is that the barrier for battery swapping is exclusivity at the technical standard level, while the barrier for charging is stickiness at the network density and user experience level.
It is worth noting that battery swapping and ultra-fast charging are considered to coexist long-term rather than replace each other.
Yang Xi emphasizes that battery swapping has already achieved market education and formed a fixed user base through NIO, achieving rapid replenishment just like megawatt-level flash charging; the two technical routes support grid dispatch and energy trading with consistent logic, ultimately aiming for the same goal: making EV energy replenishment as fast and cheap as refueling gasoline cars.
This means automakers' investments in energy replenishment will not bet solely on a single technical route.
Networks Denser, Standards Set: When Will the Profit Model Be Proven?
However, the expansion of energy replenishment networks does not imply that the business model has been proven viable.
According to Yang Xi, the entire industry is currently still in the investment phase of infrastructure deployment. Players prioritize network coverage, and exploration of commercial profitability models has not truly started. At least three issues confront the industry.
First, the ecosystem is not yet formed, and profit models are still being explored.
The major cost of battery-swap stations lies not in initial infrastructure but in subsequent operation, maintenance, and electricity fees. Each station has a clear break-even point but requires sufficient traffic flow to fill it. The same applies to charging networks. Reportedly, BYD's flash-charging stations achieve 2,983 kWh of daily charging volume per gun, but this is the result after dilution across the scale of 10,000 stations.
It can be judged that a mature energy replenishment commercial ecosystem still needs a considerable amount of time to form. Currently, players are using "scale to buy time," betting that network density will first support the profit model.
Second, the battle over standards has intensified to a white-hot state.
Battery-swap standardization is the biggest bottleneck for scaling. Before the Geely-NIO cooperation, the two companies had two separate battery-swap standards. In this cooperation, Geely chose to co-build unified C-end battery-swap technology and standards with NIO, rather than insisting on independent standards.
NIO's battery-swap standards are relatively mature. It previously led the formulation of national industry standards for the National Energy Administration, such as General Technical Requirements for Locking Mechanisms of Quick-Replacement Battery Boxes for Electric Vehicles and General Technical Requirements for Cooling Interfaces of Quick-Replacement Battery Boxes for Electric Vehicles; CATL attracts automakers into the same ChocolateSwap standard through its "Super-Swap Shared Network," planning to build over 100,000 shared facilities by the end of 2028.
Once standards are established, it implies long-term rule-making authority.
Third, the dilemma of balancing openness and closedness.
Self-built networks are closed and controllable but costly with limited utilization rates; open sharing is efficient but difficult to differentiate.
Tesla has opened nearly 40% of its Supercharger stations to the industry, and nearly one-third of BYD's 1.83 million flash-charging users are owners of other brands, demonstrating that openness is becoming a trend.
However, for automakers who have invested hundreds of billions of CNY in building networks, the hesitation that "it is difficult to close the loop after opening up" persists.
Yang Xi also raised a longer-term question: After the widespread adoption of flash charging, replenishment times will drop to the 10-minute range. Will energy replenishment then return to traditional gas station commercial scenarios? There is currently no consensus in the industry.
Previously, there was another less-discussed direction for energy replenishment: overseas markets.
According to Yang Xi's observation, Chinese automakers' current overseas deployment of energy replenishment is primarily "passive accommodation." Local infrastructure development overseas progresses slowly; without building stations, it is difficult to alleviate local users' concerns about purchasing. There is also no replicable path in overseas markets: constraints such as power supply, engineering qualifications, and fire safety vary by country.
For example, the Spanish grid in Europe lacks redundant power supply, so replenishment stations must rely on nighttime off-peak storage to obtain electricity; Brazil in South America supports energy replenishment construction at the policy level, with faster implementation than in Europe; the Middle East explicitly prohibits market-based energy trading, meaning energy replenishment networks can only be dominated by state entities.
These differences dictate that outbound automakers must calculate energy replenishment investments country by country.
In summary, the logic behind automakers deploying energy replenishment has changed significantly: from a supporting cost to "avoid the awkwardness of selling cars," to a user lever for "differentiated service," and then to "strategic positioning" in 2026. Energy replenishment has transformed from a cost item to a strategic item.
But the race is far from settled.
Who has the denser network, whose standards are adopted by more automakers, and who proves the profit model first will determine the final landscape of this energy replenishment competition.