French Quantum Computing Is Moving Out of the Lab

French quantum computing had a busy September, with chip manufacturing milestones, defense software projects and fresh revenue numbers all landing within weeks of each other.

Quantum still sounds like a research topic for physicists. But the latest roundup from The Quantum Insider shows something more practical: French companies are making qubits on commercial factory lines and wiring them into the same supercomputing stack that enterprises already use.

Qubits Built on a Commercial Chip Line

The most useful news for anyone who cares about scale came from Quobly. The company demonstrated qubit readout plus single and two-qubit operations on silicon chips made at STMicroelectronics’ commercial 300mm facility in Crolles. That is the same kind of fab that produces conventional semiconductors.

Why does that matter? Because the biggest question in quantum has always been manufacturing, not physics. A lab device that takes months to build by hand will never serve a bank or a logistics company. A chip that rolls off a standard line at least has a path to volume.

Quobly is also working with CEA-Leti through access to the FAMES Pilot Line to test next-generation processes. It signed an MOU with Orange Quantum Systems on testing silicon spin qubits at scale, and reached a first milestone with Absolut System on the cryogenic infrastructure these machines need to run.

Quantum Meets Classical Computing

Here is a point that gets lost in the hype: nobody expects quantum machines to replace your servers. The realistic model is hybrid, where a quantum processor handles one narrow part of a problem and classical systems do everything else.

French players are building that bridge. CEA and Alice & Bob committed to developing software that links quantum processors with traditional supercomputers. Quandela published a white paper on integrating photonic quantum processors with NVIDIA infrastructure. These are plumbing projects, and plumbing is what turns a demo into a product.

For software teams, this is the signal worth watching. When the interfaces between quantum and classical systems settle, developers will call quantum resources much like they call a GPU cluster today, through APIs and orchestration tools.

Defense Money and Commercial Focus

Defense is a major customer. Thales, Bull and CEA launched Aliquante, a five-year software platform project for programming quantum computers in classified environments. C12 and Thales announced QuantumTrack, a hybrid quantum-classical radar tracking solution.

Not everyone is following the same path, though. Pasqal ended its participation in the French defense program LSQUARE to concentrate on commercial applications. That is a notable strategic call, and it suggests the company sees a bigger market outside government contracts.

What the Revenue Numbers Say

Pasqal reported first-half 2026 revenue of 4.9 million euros, up 14% from a year earlier. It disclosed 70.4 million euros in booked and awarded business as of June 30, an operating loss of 59.2 million euros, and cash reserves of 110.8 million euros.

Read those figures with a cool head. Revenue is growing, but it is tiny next to the loss. Cash covers roughly two years at the current burn rate, which means the company needs either faster sales or fresh funding. The 70.4 million euros in booked business is the more encouraging number, since it points to customers committing ahead of delivery.

Quandela, meanwhile, signed agreements with Quanova to build quantum capabilities in Vietnam and with CMC Microsystems to offer cloud-based services through Canada’s Quantum Computing Sandbox. Cloud access is how most companies will first touch quantum hardware.

What This Means for Businesses Outside Europe

Should a mid-size company in Karachi, Manchester or Chicago start a quantum project this quarter? Probably not a full one. But there are low-cost moves that make sense now.

Start with awareness: pick one or two problems in your business that involve heavy optimization, such as routing, scheduling or portfolio balancing, and note them. Next, make sure your data and infrastructure are clean enough to feed any advanced compute service later. A messy data estate blocks quantum just as it blocks machine learning.

Then, watch the cloud offerings. Sandboxes like the one Quandela is joining let teams experiment without buying hardware. A small pilot with two engineers and a clear success measure costs far less than waiting and then rushing.

Also keep an eye on security. Quantum progress feeds the case for post-quantum cryptography, because encrypted data stolen today could be read later. Inventory where you use long-lived encryption and plan your migration early.

A Reality Check

None of these announcements means useful, fault-tolerant quantum computers are here. Most of the news is about prototypes, partnerships and white papers. Timelines in this field slip often, and a 14% revenue rise on a small base is not a market boom.

Still, the shift from “can we build a qubit” to “can we build ten thousand on a standard fab line and connect them to a supercomputer” is real progress. It is the sort of change that looks slow until it suddenly is not.

Key Takeaways

• Manufacturing is the story: Quobly ran qubit operations on silicon made at STMicroelectronics’ 300mm facility in Crolles, pointing toward volume production.
• Hybrid is the model: CEA, Alice & Bob and Quandela are building links between quantum processors and classical supercomputers.
• Defense funds the early work: Aliquante and QuantumTrack show government and defense demand, though Pasqal is stepping away from that route.
• Revenue is early: Pasqal posted 4.9 million euros in H1 2026 revenue against a 59.2 million euro operating loss.
• Prepare cheaply: Clean your data, list optimization problems, and try cloud sandboxes before committing real budget.

How TecniForge Can Help

At TecniForge, we help businesses navigate these technology shifts. Whether you need custom software development, AI integration, or cloud migration, our team builds scalable solutions. Talk to our experts.

So tell me: which problem in your own business is hard enough that a quantum pilot might one day be worth the experiment?