Installed lithium-ion cell nameplate capacity reached approximately 2,950 GWh worldwide by year-end 2025, with another 1,100-1,400 GWh of announcements tracked for 2026 commissioning, lifting theoretical nameplate above 4,000 GWh before utilisation derate.
China-based cell lines account for an estimated 75-80% of that 2,950 GWh nameplate, with South Korea, the U.S., Japan, and a German-Swedish-Spanish cluster (Northvolt, PowerCo, ACC) holding the next tier. Cell-chemistry split has flipped: LFP moved from 28% of 2022 shipments to roughly 48% of 2025 GWh shipped, with NCM/NCA holding near 47% and LTO/lithium-manganese/niche chemistries absorbing the residual.
Country-by-Country Nameplate and Concentration
China's operating nameplate for 2025 is reported near 2,200 GWh across CATL, BYD, EVE, CALB, Gotion, Sunwoda, REPT and SVOLT, with concentration (CR5) sitting in the 70% band; the Hubei, Jiangsu, Sichuan and Guangdong clusters each individually exceed 200 GWh of operating capacity [S1][S3]. Japan's installed nameplate is in the 70-90 GWh band across Panasonic, Murata and a long tail of small-format specialists, while U.S. operating capacity has crossed the 200 GWh threshold as Tesla Nevada, Ford BlueOval (Michigan + Tennessee), and the IRA-funded SK On/Joint Ventures lines reached ramp.
European Union operating capacity remains below 80 GWh, with Northvolt Skellefteå (Sweden) at roughly 50 GWh, ACC's two French gigafactories in early commissioning, and PowerCo's Salzgitter line in tool-in phase. India sits in the 25-40 GWh band, with Ola, Ather, Exide, Amara Raja and Tata Agratas accounting for most of the announced 80+ GWh pipeline. The 2,950 GWh global nameplate figure implies a utilisation derate to 65-72% in 2025 because of ramp curves and 2024-25 destocking; effective output is reported closer to 2,000-2,150 GWh shipped [S3].
Cell-Chemistry Mix: LFP Overtakes Ternary in 2024-2025
NCM/NCA at 47% remains the dominant chemistry in Korean, Japanese and U.S. automotive programs, with cell-level energy density 5-15% above LFP at the cost of higher nickel/cobalt price exposure.
Within NCM, the migration from 5-2-2 to 6-1-1 and 8-1-1 ratios has moved 2025 cell energy density into the 250-300 Wh/kg range at the cell level, with pack-level at 200-230 Wh/kg. LTO (lithium titanate) holds less than 2% of global GWh shipped, used in trolleybuses, mining locomotives, and special industrial vehicle applications where 15,000+ cycle life at wide state-of-charge window is specified [S2]. The related steel-fibre reinforcement work parallels these high-cycle battery packs in that the fatigue/calibration data drives the operating envelope — both fields punish generic spec'ing.
Upstream Cathode and Anode Capacity Bottlenecks
Cathode-active material (CAM) nameplate is now the binding constraint, with 2025 global CAM capacity reported near 3,400-3,800 kt/yr versus a cell industry that consumes roughly 3,000-3,200 kt/yr at current mix; the bottleneck has moved from lithium chemicals to CAM conversion [S3]. LFP cathode lines are concentrated in Hunan, Sichuan, Jiangsu and Hubei (Hunan Yuneng, Dynanonic, Rongbay, German-listed BASF Shanshan), and an estimated 70-80% of LFP CAM capacity sits in China.
Anode active material (graphite, mostly synthetic and natural spherical graphite) sees a similar concentration: an estimated 85-90% of anode capacity in China, with 2025 graphite output of 2,400-2,800 kt/yr against the 2,000-2,300 kt/yr of cell demand. Solid-electrolyte and silicon-anode pilot lines total in the low single-digit GWh-equivalent, far below the multi-GWh production lines needed to move the needle on shipments before 2027-2028.
Format Migration: Prismatic, Pouch and Cylindrical Split
Global 2025 cell shipments split roughly 56% prismatic (LFP-dominant, used by CATL/BYD and most Chinese OEMs), 32% pouch (NCM/NCA, dominant in Korean, Japanese and European programs) and 12% cylindrical (Tesla 4680, Samsung/LG 21700, EVE, BAK) [S1][S3]. 4680-format ramp has been slower than announced: total 2025 output is reported under 50 GWh globally, with Tesla Nevada and Tesla Berlin (in commissioning) as the main sites.
For warehouse and logistics spec writing, rough terrain forklift sizing bands increasingly reference LFP 280-314 Ah prismatic cells; the cell-format choice drives voltage-pack (96V, 144V) and pack BMS architecture decisions. Similarly, order picker spec bands for class III electric trucks now default to 48-80 V LFP packs because the cycle life at 80% DoD exceeds 4,000 cycles, lowering the 10-year cost-of-ownership versus older LTO or NMC packs.
2026 Commissioning Pipeline and Risk Bands
The 1,100-1,400 GWh of 2026 commissioning pipeline is concentrated in China (60-70%), Europe (15-20%), and the U.S. (10-15%); the largest single new site in 2026 is reported as CATL's Yibin Phase-6, with an additional 120 GWh of LFP lines [S1][S3]. Risk bands: (a) Western EV demand has cooled versus 2023 forecasts, with 2025 global BEV+PHEV sales reported near 17-18 million units, putting utilization pressure on 2024-2025 commissioning; (b) tariff regimes have not aligned: U.S. Section 301 7.5-25% duties on Chinese cells push Korean/Japanese supply; (c) Europe's Carbon Border Adjustment Mechanism (CBAM) reporting on battery cells takes effect progressively from 2026.
For sourcing teams, the practical 2026 question is not "is there capacity" but "is the cell UL 1973 / IEC 62619 / UN 38.3 / IATF 16949 certified for the specific format and chemistry." The Chinese LFP 280 Ah prismatic cell, for example, is widely available in 16-22 week lead-times [S1], while the same 280 Ah format with IATF 16949 audit at a non-Chinese plant typically runs 32-44 weeks. The mismatch in lead-time is the actual buying decision for stationary storage and light-industrial mobility buyers in 2026.
Industrial and Stationary-Storage Off-Take
Stationary energy storage system (ESS) deployments have been the swing factor for 2024-2025 LFP utilisation, with 2025 ESS GWh shipped reported in the 220-260 GWh range, a 60-90% year-on-year increase depending on source [S3]. The 314 Ah LFP prismatic cell has become the de-facto ESS spec in Chinese, U.S. and Middle-Eastern tender documents because of its 8,000-10,000 cycle rating at 80% DoD and 25 °C operation.
Industrial-vehicle battery packs — forklifts, AGVs, port equipment, mining trucks — represent a smaller but high-margin slice. Frey Battery, a Chinese industrial-vehicle specialist with 16 years of LFP pack design, ships 48-96 V packs in the 200-700 Ah range for mining and special-purpose vehicles, with operating temperature windows of -20 °C to +60 °C reported for cold-climate coal-haul deployments [S2]. For material-handling buyers, the rough-terrain forklift sizing logic ties to LFP cell availability and UL/UN38.3 pack certification; an underspecified pack is a field failure, not a warranty issue.
Sourcing Watch-List Through 2026 Q4
Four trackable signals for 2026-2027 capacity allocation: (1) CATL Yibin Phase-6 and Phase-7 commissioning dates and first-customer allocations — these lines alone will add roughly 200-240 GWh of LFP nameplate, materially shifting the 2027 supply curve; (2) Northvolt Skellefteå ETT-cell line conversion — the 2026-2027 shift from wound to Z-folding will determine whether European prismatic LFP supply is competitive with Chinese imports; (3) U.S. Section 301 exemption status on Korean-built cells — every 5 percentage points of duty change moves the U.S. effective landed cost by USD 4-7/kWh on a 2026 NMC-811 prismatic cell; (4) LFP CAM export licences out of China — the policy posture in H2 2026 will determine whether Korean/Japanese LFP cell lines can source cathode outside their existing Chinese offtakes. [S1]
Trackable market signal: the 2026 BloombergNEF Lithium-Ion Battery Price Survey (annual July release) is the next hard data point on cell ASP and on whether the 2024-25 USD/kWh decline resumed after the 2024 destocking trough.
For component-level specifications, see pressure transmitter, flow meter, and industrial valve.