Chinese electric-vehicle makers are accelerating a shift away from their heavy reliance on CATL, the world's largest battery maker, as the supply chain evolves from deep, exclusive partnerships to a more diversified structure. According to a report from Huxiu, automakers are not severing ties with CATL but are changing how they work with it: moving from a single-supplier, black-box model to a white-box approach where they define cell specifications and open up to multiple vendors.
From scarcity to surplus
The change is driven by a dramatic reversal in market conditions. In 2021-2022, lithium carbonate prices soared from tens of thousands of yuan per tonne to hundreds of thousands, and cell capacity was severely tight. Automakers queued for supply, and premium capacity was reportedly sold out through 2025-2026. Deep binding became standard practice: automakers signed long-term supply agreements with CATL, sometimes prepaying or co-investing in plants. CATL, in turn, supported upstream suppliers with capital and purchase commitments, such as taking a stake in Hunan Yuneng, which grew from a second-tier player to the world's top lithium iron phosphate (LFP) cathode producer.
But this model fueled an overcapacity wave. Around 2022, more than 50 companies announced expansion plans totaling over 1.4 trillion yuan (about $197 billion), with planned capacity reaching 2,500 GWh—far exceeding market demand. Capacity utilization in China's power battery industry fell from 51.6% in 2022 to 41% in 2023 and is projected at only about 35% for 2025, according to the China Automotive Power Battery Industry Innovation Alliance. However, high-end capacity remains tight: CATL's utilization rate returned to above 95% in the second half of 2025, with orders booked into the following year.
Market demand has also shifted. LFP's share of installations rose from 51.7% in 2021 to 81.2% in 2025, as the market prioritized cost and safety over extreme energy density, thinning the technology moat. From 3C to 5C fast charging, BYD, CALB, EVE Energy, and Sunwoda have progressed in step, and manufacturing techniques like lamination, AI defect detection, and X-ray inspection are now common among leading battery makers.
Automakers take control
In this context, automakers are turning battery procurement into a portfolio of options. Li Auto has taken an equity stake in a second supplier: in September 2026, it invested 2.65 billion yuan (about $373 million) to become the second-largest shareholder of Sunwoda Power. Its self-developed 5C ternary lithium batteries are being phased into models from the L8, L6, and i8 to the MEGA and i9, though the i9's first deliveries still use CATL cells until self-production ramps up.
Xiaomi is leading cell definition for its Longjia battery, with CALB and Sunwoda supplying cells while Xiaomi controls specifications, pack structure, and BMS software, using a "shadow factory" mechanism for quality control. Xpeng has announced it will develop and manufacture its own battery systems, but explicitly not cells. The AITO brand, long exclusively supplied by CATL, is now bringing in CALB and Gotion High-tech as second and third suppliers, both entering Huawei's "Whale" battery supply chain. Leapmotor is procuring from five suppliers in parallel, while Nio's sub-brand Onvo has added CALB and FinDreams.
CATL itself is also restructuring. Its supply agreement with Hunan Yuneng expired in 2024 and was not renewed; since September 2025, it has signed long-term LFP purchase agreements with Ronbay Technology (3.05 million tonnes for 2026-2031) and Fuling Precision (at least 3 million tonnes), and it has reduced its stake in Hunan Yuneng, cashing out at least 1.1 billion yuan (about $155 million) by September 2026. CATL is also moving further upstream, investing 30 billion yuan (about $4.2 billion) in April 2026 to establish Times Resources Group, a mining investment platform, and 1.4 billion yuan (about $197 million) in a lithium salt project in Ya'an, plus 8.5 billion yuan (about $1.2 billion) to acquire Chongqing Yaoning from Geely.
Second-tier cell makers are building independent supply chains. EVE Energy has joint ventures with material suppliers like BTR, Huayou Cobalt, Enjie, and Capchem. CALB has secured a supply commitment for 725,000 tonnes of electrolyte from Tinci Materials for 2026-2028 and is building its own 100,000-tonne cathode project. At the material level, Hunan Yuneng's sales to CATL fell from 42.82% in 2022 to 29.73% in 2025, and its top-five customer concentration dropped from 96.11% in 2023 to 74.79% in the first three quarters of 2025, adding customers like EVE Energy and Envision AESC.
Industry standards are also pushing change. The mandatory national standard GB 38031-2025, effective July 2026, makes "no fire, no explosion" a legal requirement rather than a selling point. Automakers with self-developed batteries are even tracing standards down to second- and third-tier supply chains, imposing consistency requirements on smaller battery makers that match those of top players. While some material suppliers like Enjie and Xingyuan in the separator segment already have diversified customer bases—the top four held 72% of the separator market in 2025—others are struggling: Dynanonic's market share fell from 19% in 2021 to 7.48% in the first half of 2026.
Whether automakers can convert their new definition power into product and cost advantages, and whether second- and third-tier suppliers can deliver on consistency, yield, and volume, will determine the depth and pace of this industry restructuring.
