[SMM Analysis]Automakers Accelerate In-House Battery Programs: A Power Shift from "Buying Cells" to "Defining Cells"

Recently, Huawei unveiled its Whale battery platform, the all-new Li Auto L8 debuted with a Li Auto-led self-developed 5C battery pack, and Xiaomi officially launched its Longjia ("Dragon Armor") battery — a string of major battery announcements from automakers in quick succession. Unlike the traditional model in which carmakers simply purchased standardized cells from battery manufacturers and handled only vehicle integration, automakers are now deeply involved in every stage of the process: battery definition, material selection, development validation, and quality control. In-house battery development is evolving from a technical experiment at a handful of companies into an industry-wide strategic wave.

Three Models, Three Paths

The first is full in-house development and production, with BYD's FinDreams Battery as the only proven success story — vertical integration spanning materials, cells, and vehicles. Yet this model demands annual sales volumes in the millions to sustain it. As Liu Kai, Executive Secretary-General of the NEV Battery Branch of the China Association of Automobile Manufacturers, puts it bluntly: "Strictly speaking, BYD is the only company that has truly made this path work."

The second is the "automaker defines, battery maker manufactures" joint-venture model, for which Li Auto and Sunwoda have provided the industry's first complete case study. The two sides established a 50:50 battery joint venture, with Li Auto leading full-stack fundamental development of cell materials, structural systems, and BMS — backed by cumulative investment exceeding RMB 500 million and a battery team of nearly 300 people — while Sunwoda provides its mature manufacturing platform. The all-new Li Auto L8, launched in June 2026, is equipped exclusively with this self-developed battery, with CATL exiting supply for the model. Li Auto has also announced that from 2026 onward, its entire lineup will carry batteries from only two brands: CATL and its own "Li" brand. Where automakers once fought for the right to define battery systems, Li Auto is now fighting for product rights over the cell itself.

The third is group-level consolidation for in-house supply, exemplified by Geely. The company consolidated its battery assets into a single entity, Jiyao Tongxing, whose Aegis "Gold Brick" batteries exclusively supply Zeekr, Lynk & Co, and Galaxy. Geely has also formalized a long-term supply structure of "40% in-house + 40% CATL + 20% others" — in-house production as the foundation, not the entirety.

EV batteries account for roughly 30-40% of a vehicle's total cost, making them the single largest variable in automakers' profitability. During years of violent lithium price swings, carmakers passively absorbed battery price hikes and suffered from an imbalanced distribution of profits across the value chain, making supply chain control a necessity. Meanwhile, battery performance — fast charging, safety, low-temperature resilience — has increasingly become a core selling point, meaning that control over cell definition equates to control over product competitiveness. As Zhang Chengyun, Director of International Consulting at China Automotive Engineering Research Institute, notes, even though lithium carbonate prices have retreated significantly from their previous peak of RMB 600,000 per tonne, the imbalanced profit distribution between upstream and downstream players remains fundamentally unresolved — and that is what is driving more automakers to commit.

For Battery Makers: A Double-Edged Sword Unsheathed

For second-tier battery manufacturers, the in-house battery wave brings both orders and concessions. Take Sunwoda: the contract-manufacturing model delivers assured capacity utilization and binding relationships with leading clients — beyond Li Auto, it supplies 60% of the battery packs for Xiaomi's new sub-brand and has entered Huawei's HIMA (Harmony Intelligent Mobility Alliance) supply chain. The trade-off is ceding cell definition rights and a portion of per-vehicle value, leaving the company primarily with manufacturing capabilities. That said, cell design and manufacturing are deeply coupled — materials, processes, and yield rates cannot be cleanly separated — which means the contract model will not upend battery makers in the near term, but it will reconfigure how profits are shared.

The industry landscape is diverging. Leading battery makers have already responded: CATL listed "intensifying market competition" as a risk factor in its interim report while accelerating its consumer-facing branding — "Choose CATL for your EV" advertisements now appear in airports and high-speed rail stations, and its promotional spending nearly doubled in the first half of the year — in a bid to embed its brand directly into consumers' minds. The battery industry can be expected to split into two distinct species: "brand cell makers" that command technology and brand definition, and "contract manufacturing platforms" that fulfill automakers' in-house battery orders. For second-tier players, becoming a capacity partner in the latter camp is not a retreat but the most realistic survival strategy amid structural overcapacity — the key question is whether they can defend their quality moats and cost advantages in manufacturing while giving up definitional control. This transfer of power has only just begun.