Quobly Runs Three Quantum Operations on a Commercial Chip Line

Quobly said on Sept. 16, 2026, that it had run three basic quantum operations, qubit readout, a single-qubit gate and a two-qubit gate, on a single chip fabricated on STMicroelectronics' commercial 300-millimeter production line in Crolles, France, using the fully depleted silicon-on-insulator, or FD-SOI, process the same line uses for ordinary logic chips. "Having these three basic operations demonstrated on a single QSOI chip is an important step," said Tristan Meunier, Quobly's chief scientific officer and co-founder, according to the company's announcement.

The number Quobly did not publish

What the announcement does not say is how good those operations were. Quobly gave no error rate, no gate fidelity and no qubit count for the demonstration chip, stating only that "further details and performance metrics will be reported in a forthcoming scientific publication." That is the figure that actually determines whether a spin-qubit platform is competitive: a two-qubit gate fidelity in the high 90s can support error correction, while one a few points lower cannot, and the difference is invisible in a press release that only confirms the operation ran. Daniel Loss, president of Quobly's scientific advisory board, framed the milestone in process terms rather than performance terms: "Readout, single-qubit and two-qubit gates on one chip from a 300 mm FD-SOI process is the step that moves the spin qubit from the laboratory into the process flow," he said in the same announcement. That is a real claim, and a narrower one than it might first read: it says the operations survived contact with a commercial fab, not that they are good enough to build a computer from yet.

STMicroelectronics is landlord, fab and investor at once

Quobly is not renting capacity from a neutral foundry. STMicroelectronics was one of three lead investors, alongside French public investment bank Bpifrance and the semiconductor-security firm SEALSQ, in the €115 million Series A that Quobly closed on June 3, 2026, according to The Quantum Insider. That round also drew the European Innovation Council, industrial-gas group Air Liquide's venture arm and existing investor Innovacom, on top of earlier backers CEA, CNRS, Quantonation and Supernova Invest, who funded a €19 million seed round between 2023 and 2025. "This financing marks a transition from technology validation to industrial execution," Chief Executive Maud Vinet said when the round closed. The chip material itself, isotopically enriched silicon-28, is supplied through a partnership with wafer maker Soitec, and gas group Orano is also named as a supply-chain partner, so the same small group of names shows up as investor, supplier and manufacturer across the stack.

Milestone	Date	Detail
Seed round	2023-2025	19 million euros
Series A	June 3, 2026	115 million euros, led by Bpifrance, SEALSQ and STMicroelectronics
Three-operation chip demo	Sept. 16, 2026	Readout plus single- and two-qubit gates, no fidelity disclosed
Alloy Pioneer cloud access	Target: end of 2026	First commercial system, per Quobly's roadmap

Cryogenic control is the other half of the claim

The operations count is only one part of what Quobly says it proved. The company also reported co-integrating the quantum circuits with cryogenic control electronics on the same die, and it has built what it calls a cryogenic process design kit, the standard library of building blocks a chip designer uses to lay out a circuit, adapted to run at the near-absolute-zero temperatures spin qubits need, according to Quantum Computing Report. That matters because every qubit that needs its own dedicated wire running out of a dilution refrigerator is a qubit that will not scale past a few hundred; control electronics that live on the same chip, at the same temperature, are the standard answer the field gives for why silicon spin qubits could in principle pack far more of them into a given wiring budget than platforms that pipe every control line in from room temperature. Quobly's Sept. 16 statement does not say how many control channels the cryogenic PDK supports today, only that it exists.

What would change the read

Quobly's own roadmap gives two dates to hold it to. The company says it plans to open cloud access to Alloy Pioneer, its first commercial quantum system, by the end of 2026, and its longer-range target is one million qubits by 2032, according to Quantum Computing Report. Neither date depends on today's announcement alone. The variable that would change this read is the promised follow-up paper: if it reports two-qubit fidelities in the high 90s at the density a 300mm line can produce, the manufacturing story and the performance story converge, which is the combination competing platforms built on superconducting circuits or trapped ions have struggled to claim simultaneously. If the paper instead shows fidelities well short of that bar, the Sept. 16 announcement will have been a manufacturing result dressed as a computing one. A placed date on a roadmap is not a demonstrated one, the same gap that separates SpaceX's Sept. 22 target for Starship's first full orbital attempt from an actual flight. Quobly has not said when its own publication will appear, and quantum-focused capital is not waiting to find out: Matter Venture Partners closed a $450 million second fund the same day, Sept. 16, 2026.