ProLogium's solid-state cell passed the tests. Its factory makes 0.5 GWh a year

There is a particular kind of battery announcement that arrives every few months, and the honest reaction to most of them is to wait. ProLogium's is different in one respect that matters: an outside lab put a number on the cell, a second lab checked whether the thing deserves the words "all-solid-state" at all, and the company says the line is running rather than being planned.

The Taiwanese firm said on 2 September that its Generation 3.5 Lithium Ceramic Battery is in mass production at its facility in Taoyuan. A third-party TUV test report measured the 185.4 amp-hour large-format pouch cell at 381 watt-hours per kilogram gravimetric and 903 watt-hours per litre volumetric. For scale, the nickel-manganese-cobalt cells in most electric cars sold in the United States top out around 300 Wh/kg, and the lithium-iron-phosphate cells spreading through cheaper models sit between roughly 150 and 200 Wh/kg.

The more interesting document is the second one. UL Solutions tested the same cell under GB/T 43568-2026, the Chinese methodology introduced in July specifically to separate genuine all-solid-state cells from packs that quietly retain some liquid electrolyte. The cell sat for six hours under continuous vacuum at 120C. ProLogium reported mass loss below 0.05 per cent, comfortably under the standard's 0.5 per cent ceiling.

That threshold exists because the phrase "solid-state" has been doing a great deal of unpaid marketing work. CarNewsChina, which covered the announcement on 6 September, pointed to SAIC-backed IM Motors promoting a QingTao-developed pack as a "Light-Year Solid-State" battery while it still contained liquid electrolyte. A vacuum chamber does not care about branding. Whatever else is arguable about ProLogium's Gen 3.5, its classification now rests on a published method rather than a press kit — and China has submitted that method to the International Electrotechnical Commission as a reference for international standardisation.

ProLogium describes Taoyuan as a "Giga-level" facility. Its initial operational capacity is 0.5 GWh.

At 80 kWh per pack, that is roughly 6,000 to 6,250 electric cars a year. A single mainstream cell plant of the sort CATL or LG Energy Solution operates measures its output in tens of gigawatt-hours. The company's latest investor roadmap projects Taoyuan reaching 1.05 GWh by 2030, with up to 3 GWh at full design capacity. A second plant in Dunkirk, France is designed for 4 GWh by 2030 and up to 44 GWh at full design capacity, with its first 0.8 GWh phase due in 2028. The French first phase alone would be eight times what Taoyuan makes today.

So the correct reading is not "solid-state batteries have arrived in cars". It is that a company has moved a high-energy-density cell out of the pilot line and onto a commercial one, at a volume that would supply a niche model rather than a lineup.

ProLogium says cumulative LCB shipments have passed 2.4 million cells since commercial production began in 2013. That figure spans consumer, specialty and automotive applications, and it is worth reading the automotive part closely. The disclosed business includes supply to a US automotive audio-system company for vehicles built by a Japanese automaker in North America, with cumulative deliveries above 900,000 cells across more than 175 repeat orders. The company also reports around 10,000 automotive battery samples.

Repeat orders over years are meaningful evidence of process control and yield — the boring competences that kill most battery startups. They are not evidence of a high-voltage traction programme. They are not evidence that any production EV uses these cells as its traction battery.

Gen 3.5 uses what ProLogium calls its Logithium cell architecture, designed in 2012: a ceramic separator combined with a proprietary edge-frame structure that forms an additional separator around the electrode perimeter, isolating potential burrs while sealing and insulating the cell. The company has moved through three manufacturing generations to get here — sheet-by-sheet in 2013, roll-to-roll in 2017, the current Giga-level platform in 2024 — and holds IATF 16949 automotive quality certification obtained in 2022.

For the previous Gen 3 cell, ProLogium claims a 5 to 80 per cent charge in eight and a half minutes, and says the cell will not ignite when shot with a bullet, held at 170C, or overcharged to twice rated voltage. Those are company figures, not third-party ones.

Two things stand between all of this and a car. The first is that cell-level energy density is not pack-level energy density: structural members, cooling, contactors and battery-management hardware all add mass and volume, and the gain a driver experiences is whatever survives that. The second is cost. Mass-market LFP cells in China have been reported at roughly 300 to 400 yuan per kWh — about 42 to 55 US dollars — and ProLogium has not disclosed a comparable production cost for Gen 3.5.

Mercedes-Benz holds a technology cooperation agreement with the company, which has also worked with Nio and Aiways on solid-state development and sample vehicles. None of them has announced a production car built around the cell. Gen 4, with a fully inorganic electrolyte system, is next, and ProLogium says a Gen 3.5 line needs only about 10 per cent of its equipment upgraded to build it.

That is a genuinely clever piece of industrial planning — provided somebody orders enough cells to justify the line in the first place.