In China, "solid-state" is three different products sharing one marketing word. Hybrid solid-liquid (semi-solid) packs have already shipped in limited volumes. All-solid-state cells are still mostly pilots, roadmaps, and national standards work. Confusing those buckets is how buyers and investors misread 2027 slides.
This guide covers China solid-state batteries as of September 2026: how GB/T and industry drafts separate liquid, hybrid solid-liquid, and all-solid-state; what BYD and CATL have actually said about sulfide timelines; what CATL's "condensed matter" product is (and is not); which semi-solid packs shipped and then stalled; the September 2026 MIIT five-year plan and Seoul Economic Daily counter-signal; and a short skeptic sidebar on the Donut Lab controversy. For ceramic-separator licensing, see Inside Deep Tech's QuantumScape QSE-5 full guide. For quasi-solid and sulfide OEM paths into Mercedes and Stellantis, see the Factorial Energy FEST and Solstice full guide. For ceramic Logithium cells, see the ProLogium Gen4 full guide. For Toyota and Idemitsu's sulfide factory path, see the Toyota Idemitsu sulfide full guide. Numbers below come from company-executive reporting, Chinese standards pages, MIIT plan coverage, and named trade press, not invented datasheets.
Key Takeaways
- China's GB/T 48093.1-2026 (published Aug 28, 2026; implements Jan 1, 2028) defines EV solid-state terms and classes (NDLS).
- BYD CTO: demonstration use ~2027; large-scale adoption after 2030; sulfide; 20 Ah / 60 Ah trial cells in 2024 (CnEVPost).
- CATL chairman: technology at level 4 of 9; million-vehicle installs before 2030 unlikely (CnEVPost).
- MIIT five-year plan targets initial large-scale all-solid-state use by 2030 (CnEVPost).
- NIO 150 kWh WeLion semi-solid: series in Apr 2024, stopped after a few hundred packs (electrive).
When "solid-state" means three things
Here's the problem. Chinese marketing often says "solid-state" for packs that still move ions through liquid or gel. Buyers then treat a shipping semi-solid pack as proof that all-solid-state mass production has arrived. It has not.
China's national standard GB/T 48093.1-2026, Solid-state battery for electric vehicle - Part 1: Terms and classification, was published on August 28, 2026 and takes effect on January 1, 2028. Drafting units include CATARC, CATL, BYD Lithium Battery, WeLion, QingTao, SAIC, and many OEMs, which signals industry buy-in even before the effective date.
Industry working drafts that feed that taxonomy split batteries by how ions travel between electrodes. A widely circulated working draft defines three classes: liquid-state (liquid electrolyte only), hybrid solid-liquid / semi-solid (liquid and solid electrolytes both carry ions), and solid-state / all-solid-state (solid electrolyte only). That is the cleanest way to read Chinese press releases in 2026.
| Class (English) | Ion transport | What it usually means in 2026 China |
|---|---|---|
| Liquid-state | Liquid electrolyte only | Mainstream LFP / NCM packs |
| Hybrid solid-liquid (semi-solid) | Liquid + solid electrolytes | Shipped "solid-state" marketing packs; gel/polymer hybrids |
| All-solid-state | Solid electrolyte only | Pilot / demo / 2027-2030 roadmaps; sulfide or oxide paths |
Source: GB/T 48093.1-2026 NDLS record; class language from the terms-and-classification working draft.
Of course, English headlines still flatten those three labels into one word. When you see "China launches solid-state EV," check whether the cell still needs liquid or gel. That single check separates shipped hybrid packs from all-solid-state demos.
BYD: sulfide demos in 2027, scale after 2030
BYD is the world's largest NEV maker and the second-largest power-battery maker. In 2024, BYD's power-battery installed capacity was 153.7 GWh with a 17.2 percent global share, behind CATL at 37.9 percent (CnEVPost). That scale matters: BYD can fund a sulfide program without pretending the Blade LFP fleet is obsolete tomorrow.
On February 15, 2025, BYD battery-business CTO Sun Huajun told an all-solid-state battery forum that BYD expects to begin "demonstration use" of all-solid-state batteries around 2027, while true large-scale adoption may not be realized until after 2030 (CnEVPost). Material choice is sulfide, framed around cost and process stability versus oxide and polymer routes.
Sun's staged history is concrete. BYD started all-solid-state R&D in 2013. By 2016, prototypes were soft-pack cells under 1 Ah. By 2023, the company said it was validating industrialization feasibility. By 2024, trial products of 20 Ah and 60 Ah solid-state cells had rolled off a production line (CnEVPost). That is industrial intent, not a named production BEV.
On cost, Sun argued sulfide electrolyte prices look high mainly because volume is low, and that after a certain scale, solid-state ternary cells could theoretically approach liquid ternary cost levels (CnEVPost). Treat that as a company thesis, not a shipped BOM.
| BYD milestone | What was claimed | Status for readers |
|---|---|---|
| 2013 | ASSB R&D start | Historical R&D |
| 2016 | Soft-pack prototypes <1 Ah | Lab / feasibility |
| 2024 | 20 Ah and 60 Ah trial cells | Pilot line, not mass BEV |
| ~2027 | Demonstration / vehicle application | Company target |
| After 2030 | True large-scale adoption | Company target |
Source: CnEVPost reporting of Sun Huajun's February 2025 remarks.
CATL: condensed matter ships the hybrid story; ASSB stays cautious
CATL runs two narratives that get mixed in English coverage. One is condensed matter / Qilin Condensed: a high-energy product that still sits in the hybrid / semi-solid family. The other is all-solid-state sulfide, where CATL's own chairman keeps talking the industry down from 2027 hype.
Condensed matter is not all-solid-state
At CATL's April 21, 2026 Super Tech Day, the company unveiled the Qilin Condensed Battery at 350 Wh/kg gravimetric and 760 Wh/L volumetric cell energy density, with pack weight under 650 kg and claimed sedan ranges up to 1,500 km (Charged EVs). CATL says related condensed technology was validated in electric aviation at 500 Wh/kg on 4-tonne aircraft. Trade coverage is clear that this replaces conventional liquid electrolyte with a condensed system. That is progress. It is still not the all-solid-state cell Zeng scores at level 4.
Level 4 of 9, and why 2027 is not mass production
At the June 2026 Summer Davos forum in Dalian, CATL chairman Robin Zeng said solid-state technology is only at level 4 of 9, where level 9 means the batteries "can be mass produced" (CnEVPost; Electrek). Mass production, he said, needs a technology path, a product path (supply, reliability, safety), and a commercial path (whether buyers accept the price).
Zeng has also said mass adoption in millions of vehicles is unlikely before 2030. CATL chief scientist Wu Kai said in 2024 the company aimed to reach level 7 to 8 by 2027, enabling small-batch all-solid-state production while large-scale production would still face cost issues (CnEVPost). Electrek reports CATL still points to about 5 GWh of small-scale capacity for that window, plus pilot cells at 500 Wh/kg (Electrek). CATL had expanded its solid-state R&D team to more than 1,000 people by the end of 2024 (CnEVPost).
That means: treat CATL's 2027 line as a small-batch industrial milestone, and treat 2030 as the company's mass-production honesty check. Condensed matter marketing does not erase that distinction.
What has actually shipped: semi-solid, not all-solid
Let's start with the clearest customer-facing China example. In April 2024, NIO began series production of 150 kWh packs using semi-solid gel-like polymer electrolyte cells from partner WeLion, with a claimed CLTC range around 1,000 km (electrive). That is a real hybrid pack that left a factory.
It also failed commercially. By November 2025, CEO William Li said NIO had discontinued the pack after a few hundred units because customers were not swapping into it often enough and the subscription fee was too high, especially once NIO had more than 3,500 swap stations and 97 percent of users preferred 75 kWh over 100 kWh packs (electrive). Semi-solid can ship and still lose to cheaper liquid packs plus infrastructure.
Electrek's June 2026 CATL piece also notes other China hybrid examples in the market (IM "Lightyear," SAIC MG4 semi-solid volumes). Use those as hybrid shipping evidence, not as proof that all-solid-state has cleared Zeng's level-9 bar (Electrek).
| Program | Class | Public checkpoint (as of sources cited) |
|---|---|---|
| NIO 150 kWh / WeLion | Semi-solid (hybrid) | Series Apr 2024; stopped after a few hundred packs |
| CATL Qilin Condensed | Condensed / hybrid family | 350 Wh/kg / 760 Wh/L claims at Apr 2026 Tech Day |
| BYD sulfide ASSB | All-solid-state (target) | 20/60 Ah trials 2024; demo use ~2027; scale after 2030 |
| CATL sulfide ASSB | All-solid-state (target) | Level 4/9 mid-2026; small-batch path ~2027; mass ~2030 |
Source: electrive; Charged EVs; CnEVPost; CnEVPost.
Policy push vs manufacturing cold water
On September 28, 2026, CnEVPost reported that China's Ministry of Industry and Information Technology (MIIT) and six other agencies released a battery industry development plan for 2026 through 2030 (document dated September 14). The plan aims for initial large-scale use of all-solid-state batteries by 2030, plus breakthroughs on ionic conductivity, cycling stability, interfacial contact, pressure-system engineering, solid-electrolyte scale-up, and equipment such as isostatic presses (CnEVPost). It also sets goals for long-life lithium batteries to reach 15,000 charge-discharge cycles and for leading makers to hit parts-per-billion defect rates.
Three days earlier, on September 25, 2026, Seoul Economic Daily published a counter-signal: CATL chief manufacturing officer Ni Jun reportedly told an internal event that solid-state batteries are very unlikely to be commercialized in the near term and that large-scale commercialization will take at least five more years. The same piece restates Zeng's level 4 of 9 maturity score, says sulfide-based cells cost three to five times more than conventional lithium-ion, and quotes the view that installs in as many as 1 million vehicles before 2030 are extremely slim.
That is not a contradiction if you keep the buckets separate. MIIT is setting a 2030 initial large-scale policy target. CATL executives are saying the hard manufacturing and cost paths are still years out. Both can be true while 2027 remains a demo and small-batch year.
Skeptic sidebar: Donut Lab is not a China program
Why include a Finnish startup here? Because viral "production-ready solid-state" claims pollute every regional guide, including China coverage. In June 2026, Electrek summarized a Ziroth investigation arguing Donut Lab's CES-era "miracle" solid-state cell behaved like a high-nickel lithium-ion pouch, with energy density around 298 Wh/kg versus a claimed 400 Wh/kg and no verified 100,000-cycle life. Treat that episode as a reminder: demand CT, mass, cycle protocol, and independent chemistry evidence before you promote any non-OEM solid-state claim, Chinese or otherwise.
Honest limits
At first glance, China looks years ahead because semi-solid packs and condensed cells already have product names. Dig one layer deeper and the all-solid-state path still looks like everyone else's: sulfide interface contact, moisture control, yield, and cost multiples of 3x to 5x versus liquid lithium-ion per Seoul Economic Daily's recounting of CATL remarks.
- Naming is still being standardized. GB/T 48093.1-2026 publishes in 2026 but implements in 2028; English marketing will lag.
- BYD's 20 Ah / 60 Ah cells and CATL's 500 Wh/kg pilots are not named high-volume BEV SKUs (CnEVPost; Electrek).
- NIO's 150 kWh hybrid pack proved shipping is not the same as product-market fit (electrive).
- Third-party slides that assign BYD 400 Wh/kg or tens of thousands of 2030 vehicles without a BYD primary release are omitted here on purpose.
FAQ
Are China solid-state batteries already in mass-produced EVs?
Hybrid / semi-solid packs have shipped in limited volumes; all-solid-state has not reached high-volume series BEVs. NIO's 150 kWh WeLion pack entered series production in April 2024 and was later stopped after a few hundred units (electrive). CATL's chairman still scores true solid-state at level 4 of 9 (CnEVPost).
What is the difference between semi-solid and all-solid-state in China?
Semi-solid (hybrid solid-liquid) still uses liquid or gel electrolyte alongside solid electrolyte; all-solid-state uses solid electrolyte only. That split appears in China terms-and-classification drafts and in the GB/T 48093.1-2026 standard program (NDLS; working draft).
When will BYD put all-solid-state batteries in cars?
BYD's battery CTO has said demonstration use around 2027, with true large-scale adoption after 2030. The chemistry path he highlighted is sulfide, with 20 Ah and 60 Ah trial cells already made in 2024 (CnEVPost). No named mass-market model is confirmed in that reporting.
Is CATL's condensed matter battery all-solid-state?
No. Treat Qilin Condensed as a high-energy condensed / hybrid-family product, not Zeng's all-solid-state level-4 program. Charged EVs reports 350 Wh/kg and 760 Wh/L cell claims from CATL's April 2026 Tech Day (Charged EVs).
What did Seoul Economic Daily report on September 25, 2026?
That CATL's CMO sees large-scale solid-state commercialization at least five years out, with sulfide cells still 3x to 5x conventional cost and million-vehicle installs before 2030 extremely slim. Read the English piece for the full framing (Seoul Economic Daily).
How does China's MIIT plan line up with OEM timelines?
The 2026-2030 battery plan targets initial large-scale all-solid-state use by 2030, which matches the cautious OEM mass window more than the 2027 demo year. CnEVPost summarized the multi-agency plan on September 28, 2026 (CnEVPost).
How should readers compare China programs to QuantumScape, Factorial, ProLogium, or Toyota?
Compare electrolyte class and shipping evidence, not slogans. QuantumScape is ceramic-separator licensing; Factorial runs quasi-solid FEST and sulfide Solstice; ProLogium is ceramic Gen4; Toyota/Idemitsu is sulfide factory build-out. China adds hybrid packs that already left factories plus sulfide ASSB pilots at BYD and CATL.
Why mention Donut Lab in a China guide?
As a skeptic sidebar on unverified "production-ready solid-state" claims, not as a Chinese OEM program. Electrek's June 2026 summary of the Ziroth investigation is the cited source (Electrek). Demand independent evidence for any non-OEM miracle claim.
Next on the China solid-state watchlist: whether BYD names a 2027 demonstration vehicle and volume band, whether CATL's small-batch ASSB line clears level 7-8 with published protocols, how GB/T 48093.1-2026 shows up in product labeling before the 2028 effective date, and whether hybrid packs find a durable niche beyond prestige range SKUs. For power-electronics context on high-voltage vehicle and data-center architectures, see Inside Deep Tech's 800 VDC power distribution full guide. For the broader deep-tech company map, see the Top 100 Companies in the Deep Tech Industry 2026 list.
