7 Companies Say Their Solid-State Batteries Ship by 2028. The World's Largest Battery Maker Says They Won't.
Greater Bay Technology claims the first mass-producible all-solid-state cell at 500 Wh/kg. Toyota is building an electrolyte factory. QuantumScape demoed 844 Wh/L in a Ducati. But CATL, which makes 37% of the world's EV batteries, just dismissed its own solid-state production timeline. An original cost analysis shows the manufacturing gap is 3-5x.
One hundred and eight dollars. That is what one kilowatt-hour of lithium-ion battery pack costs today, according to BloombergNEF's 2025 price survey. Down 8% from 2024. Down 89% from 2010. In China, where overcapacity has turned battery manufacturing into a knife fight, BEV packs sell for $99/kWh, and LFP cells go for as little as $63/kWh.
This is the number that every solid-state battery startup has to beat. Not eventually. Not in a lab. At scale, on a production line, in a car that someone actually buys.
In the past six months, at least seven companies have announced they will ship solid-state EV batteries by 2028. Greater Bay Technology says late 2026. Toyota says 2027-2028. QuantumScape says its VW-backed pilot line is running. Samsung SDI is targeting 2027. Solid Power has BMW and Ford agreements. Factorial Energy has Mercedes and Hyundai. ProLogium has a factory in France.
One company is conspicuously absent from the sprint: CATL, the world's largest battery manufacturer, which publicly dismissed rumors of its own solid-state mass production timeline. When the company that controls 37% of global EV battery supply says the technology isn't ready, it's worth asking what it sees that others don't.
The A-Sample That Launched a Thousand Headlines
On April 14, 2026, Greater Bay Technology announced that its first A-sample all-solid-state battery cells had rolled off a production line in Guangzhou's Nansha district. Its cells contain zero liquid electrolyte. They survived nail penetration, extrusion, and thermal shock tests without fire or explosion. Energy density: 260 to 500 Wh/kg, depending on configuration. Fast-charge capability: 2-3C, meaning a 100 kWh pack could theoretically charge in 20 to 30 minutes.
GBT, backed by China's GAC Group, has filed over 50 patents covering the full technology chain. Its proprietary deep eutectic composite (ESC) electrolyte system blends organic and inorganic materials, sidestepping the traditional four-way chemistry race between sulfides, oxides, polymers, and halides. GBT says it is targeting GWh-level mass production by late 2026, with cells destined for GAC's Hyptec luxury EV models.
The word "A-sample" matters here. In battery development, an A-sample is the first physical proof that a design works. It comes before B-samples (design-frozen cells for vehicle testing), C-samples (production-intent cells from the final line), and D-samples (fully qualified, production-ready cells). GBT's announcement means the chemistry is viable. It does not mean a car will run on it this year.
For context: GBT installed 0.21 GWh of battery capacity in March 2026, according to the China Automotive Battery Innovation Alliance. That gave it a 0.37% market share and 15th place among Chinese battery makers. China installs roughly 57 GWh of batteries per month. Even if GBT hits "GWh-level" production of solid-state cells, it would represent less than 2% of China's monthly output.
Who Is Building What
| Company | Chemistry | Key Metric | Target Date | Stage |
|---|---|---|---|---|
| Greater Bay Tech | Deep eutectic composite | 260-500 Wh/kg | Late 2026 | A-sample |
| Toyota / Idemitsu | Sulfide | 1,000 km range | 2027-2028 | Electrolyte factory under construction |
| QuantumScape / VW | Ceramic separator (anode-free) | 844 Wh/L | 2026-2027 (field tests) | Pilot line running (Cobra process) |
| Samsung SDI | Sulfide | ~900 Wh/L | 2027 | Pilot production |
| Solid Power / BMW, Ford | Sulfide | 390 Wh/kg | 2028 | A-samples delivered to OEMs |
| Factorial Energy / Mercedes | FEST (solid polymer/ceramic) | 390 Wh/kg | 2028 | B-sample testing |
| CATL | Sulfide (condensed matter) | 500 Wh/kg (claimed) | Dismissed production rumors | R&D |
| ProLogium | Oxide (ceramic) | ~500 Wh/kg | 2028 | Factory under construction (Dunkirk, France) |
The table looks impressive until you notice the "Stage" column. No company has reached D-sample. QuantumScape (pilot line producing QSE-5 cells with a 25x-faster Cobra separator process) and Factorial (B-sample testing with Mercedes) are the furthest along by their own accounts. ProLogium broke ground on a factory in Dunkirk, France, backed by Mercedes-Benz, but that facility won't produce at scale until 2028 at the earliest. Everyone else is at A-sample or earlier.
The Cost Gap Nobody Talks About
Here is the calculation that matters most and that no press release includes.
Today's lithium-ion battery economics, per BloombergNEF:
- Average pack price: $108/kWh
- BEV packs specifically: $99/kWh
- China LFP packs: $81/kWh
- China LFP cells alone: ~$63/kWh
Now the solid-state side. Sulfide-based solid electrolytes (Li₆PS₅Cl, the chemistry favored by Toyota, Samsung SDI, and Solid Power) cost roughly $50 to $100 per kilogram in small volumes. Liquid electrolytes for conventional lithium-ion cells cost $5 to $15 per kilogram. A typical EV cell uses approximately 0.3 to 0.5 kg of electrolyte per kWh of capacity.
Just the electrolyte cost difference adds $13 to $43 per kWh to a solid-state cell, before accounting for anything else. A new low-pressure production method published in January 2026 claims a 70% reduction in electrolyte manufacturing costs. Even at 70% lower, sulfide electrolytes would cost $15 to $30 per kilogram, still 2-3x more than liquid.
But electrolyte cost is only part of the problem. Manufacturing yield is the real killer. Mature lithium-ion production lines achieve cell yields above 95%. Solid-state lines, according to industry sources and academic estimates, currently run below 50%. Every cell you throw away doubles your effective cost per good cell.
Run the full calculation. Electrolyte premium: +$15-25/kWh (assuming cost reductions). Yield penalty at 50%: effectively 2x the cell cost. Add the capital expenditure for new dry-room equipment, new stacking processes (solid-state cells can't be wound like conventional cells), and new quality control systems. First-generation solid-state packs will likely cost $300 to $500 per kWh.
That is 3 to 5 times the price of a lithium-ion pack that already works.
Why CATL Isn't Rushing
CATL's calculus is different from GBT's or QuantumScape's. CATL shipped 321.1 GWh of batteries in 2024. Its LFP cells are the cheapest and most reliable on the market. Its Shenxing and Qilin packs already deliver 600+ km of range with 4C charging (10 minutes to 80%).
For CATL, solid-state is a hedge, not a product. CATL has filed extensive patents and developed what it calls "condensed matter" battery technology with energy densities above 500 Wh/kg. But it has publicly said that mass production is years away, not months. Robin Zeng, CATL's founder, has repeatedly emphasized that safety validation, not energy density, is the bottleneck.
He has a point. A liquid electrolyte, despite being flammable, has 30 years of failure-mode data. Engineers know exactly how it degrades, shorts, and fails. Solid electrolytes are different. Lithium dendrites can propagate through grain boundaries in ceramic separators. Interface delamination can cause sudden capacity loss after thousands of cycles. These failure modes are less understood and harder to test for.
When you already sell $99/kWh packs that charge in 10 minutes and your customers are happy, the incentive to rush an unproven technology with 3-5x cost and unknown failure modes is approximately zero.
Why It Works and Why It Might Not Ship
The strongest counterargument comes from QuantumScape's Ducati demonstration. In September 2025, a Ducati V21L electric motorcycle equipped with QSE-5 cells showcased 844 Wh/L volumetric energy density, charged from 10% to 80% in approximately 12 minutes, and discharged at 10C. Not theoretical numbers. A vehicle that a person rode.
VW's battery arm PowerCo has committed up to $261 million in milestone payments to QuantumScape, extending the startup's cash runway to roughly 2029. The Cobra ceramic separator process, integrated into production in 2025, delivers 25 times the throughput of QuantumScape's previous manufacturing method. This is not a PowerPoint company anymore.
Toyota's approach is equally concrete. Idemitsu Kosan, a Japanese oil refiner turned battery materials company, began construction on a pilot plant to manufacture solid electrolytes for Toyota's packs. The factory should be complete by end of 2027, producing several hundred tons of solid electrolyte annually. Toyota's roadmap targets first-generation solid-state EVs with 1,000 km (620 miles) of range and 10-80% charging in 10 minutes by 2027-2028. A second generation aims for 1,200 km (745 miles).
But is "several hundred tons annually" enough? At roughly 0.4 kg of electrolyte per kWh, Idemitsu's factory could supply approximately 500 MWh to 750 MWh of solid-state cells per year. China installs that much lithium-ion capacity every 6 to 9 minutes. Toyota knows this. Its initial solid-state vehicles will be low-volume, high-end models, not Corollas.
What We Did Not Prove
This analysis uses publicly available cost estimates for solid-state electrolytes, which vary significantly across sources. No company has disclosed actual cell-level manufacturing costs for solid-state production. The yield figures (<50% for solid-state, >95% for lithium-ion) are composites from academic literature and industry conference presentations, not audited production data. The $300-$500/kWh estimate for first-generation solid-state packs is our own calculation based on these inputs. Actual costs could be lower if breakthrough manufacturing processes (like QuantumScape's Cobra separator or GBT's deep eutectic system) achieve yields faster than historical battery technology scaling curves suggest. The range of 260-500 Wh/kg claimed by GBT spans nearly 2x and likely reflects different cell configurations at different maturity levels, not a single product.
What You Can Do
If you're buying an EV in 2026-2027: Buy it with lithium-ion. Solid-state EVs will be luxury-tier, low-volume, and unproven in the field. LFP packs at $81/kWh already deliver 300+ miles of range, charge in under 20 minutes with 4C cells, and have a decade of reliability data behind them.
If you're investing in battery companies: Watch manufacturing yield data, not energy density claims. QuantumScape's Cobra process proving >90% yield at scale would be the single most important milestone in the entire solid-state industry. Until then, the stock trades on hope, not revenue. QS remains pre-revenue with approximately $800 million in cash and a quarterly GAAP loss of $115 million.
If you work in battery manufacturing: CATL's sodium-ion Naxtra battery (500 km range, mass production starting 2026) is a more immediate disruption to watch. It eliminates lithium and rare-earth dependency entirely. Solid-state is a 2030 story. Sodium-ion is a 2026 story.
The Bottom Line
Solid-state batteries are real. The chemistry works. Multiple companies have cells that charge fast, don't catch fire, and pack more energy per liter than anything on the market. That is not the question anymore. The question is whether they can be manufactured at a cost that competes with a $108/kWh industry that keeps getting cheaper every year. For now, the answer is no. CATL knows it. The startups racing to be first are betting that manufacturing scale will close the gap before their cash runs out. History says that bet takes longer than the press releases suggest.