Groove Quantum's EIC Accelerator Bid for Germanium Qubits

Superconducting circuits lead the field in qubit count and trapped ions in coherence, as Bits&Chips summarised it in October 2026. Groove Quantum, a Delft start-up registered in January 2024, is betting on a third route: qubits made from the spin of "holes" in germanium, in devices that its website says are each a transistor and can be built with the semiconductor industry's existing processes. The EIC Accelerator selected its project, DISQO, at the March 2025 cut-off for blended finance.

CORDIS records that Groove Quantum B.V. holds a grant agreement, signed on 12 November 2025, for an EU contribution of €2,499,875, running from 1 December 2025 to 30 November 2027. The project's stated aim is to finish developing the platform for industrial use, make it in semiconductor facilities, and begin commercial deployment through cloud access and systems installed at customers' sites. Eleven months into the project, CORDIS shows no report or result.

Why germanium, and why holes

Spin qubits store information in the spin of a single charge trapped in a tiny electrostatic cage, a quantum dot. Other start-ups build them in silicon, aiming to reuse conventional chip manufacturing, per the same Bits&Chips article. Groove's founders come from the group at QuTech, the Delft research institute of TU Delft and TNO, that pushed germanium instead: in March 2021 QuTech announced a four-qubit germanium processor arranged in a two-by-two grid, published in Nature with Hendrickx as first author, which it described as the first two-dimensional grid of semiconductor qubits.

Zwerver set out the argument to Bits&Chips in October 2026: holes in germanium have stronger spin-orbit coupling than electrons in silicon, which the company says simplifies the device architecture, and a lower effective mass than electrons in silicon. "It means we can place qubits farther apart, to a spacing of 200-250 nanometers," she said, adding that the larger spacing creates process margin. Whether that margin holds on a standard production line is the open question of the EIC project.

What has been shown so far

On 1 April 2026 Groove and QuTech authors, Zwerver and Hendrickx among them, posted a preprint on arXiv describing the simultaneous operation of an 18-qubit array in germanium, built from repeatable unit cells in what they call an extendable 2xN architecture. The abstract reports single-qubit gate fidelities averaging 99.8%, with a median of 99.9%, and controlled-Z two-qubit gates used to create a three-qubit entangled state. A preprint has not been through peer review. The company's announcement of 30 April 2026 calls the device the largest semiconductor spin-qubit processor of its kind; Bits&Chips reported this as the company's claim, and the October 2026 Bits&Chips interview quotes the company more cautiously, as matching the largest spin qubit devices shown worldwide.

The fidelity figures in the project description are higher than those in the preprint. The CORDIS objective cites single-qubit fidelities above 99.99% and two-qubit fidelities above 99.3%, and the company website lists a single-qubit result above 99.99% from a 2023 Nature Communications paper and a two-qubit result above 99% on a ten-dot array from a 2024 Science paper. Those papers predate the 18-qubit array, which the company describes as its largest device. The preprint reports the figures for the whole array operated at once, and its average single-qubit figure is lower.

1. Scale the qubit countIn progress18 qubits operated together in a 2026 preprint; 100 targeted by the end of 2027.
2. Consistent, high-performance qubitsPartly shown99.8% average single-qubit fidelity across the array, below the 99.99% cited for smaller devices.
3. Manufacturing in semiconductor facilitiesNot shownNo foundry partner or foundry-made chip announced.
4. Cloud access and on-site systemsNot shownNo commercial system or customer announced as of October 2026.

The money

When the March 2025 results were published, Silicon Canals and The Quantum Insider reported a €10 million EIC Accelerator award including a €2.5 million grant, with the remainder as EIC Fund equity, and quoted the company saying it had been chosen as one of 40 from about a thousand applicants. Neither gave the equity amount. CORDIS records the grant only.

In April 2026 the company announced €16 million: a €10 million equity seed round co-led by Innovation Industries and 55 North, with Verve Ventures and the EIC Fund taking part, and €6 million of grants from the EIC Accelerator and the Chips Joint Undertaking, as reported by The Quantum Insider and Quantum Computing Report. Pulse 2.0 puts the EIC Fund's equity at €7.5 million and says the grant is meant to take the platform from technology readiness level 6 to 8. No source breaks down how the €10 million of equity was split between investors, or how much of the €6 million came from the Chips programme, so the figures below are shown side by side, not added up.

The company's register entry, as republished by Liza from the Dutch Chamber of Commerce, shows an active private limited company at the Delftechpark address also given on its website and in CORDIS. Staff numbers are not published, and the company has announced no revenue or customer. It says it holds patent applications on qubit performance, yield and scalability, and it is recruiting.

Where it fits in EU policy

The Commission's Quantum Europe Strategy of 2 July 2025 aims to make Europe a quantum leader by 2030 and includes a quantum design facility and six quantum chip pilot lines, with a Quantum Act proposal expected in 2026. A qubit that can be made in a conventional semiconductor process links that strategy to Europe's chip policy, which is also where part of Groove's April 2026 grant money came from, per the company. CORDIS tags DISQO as 100% relevant to the digital agenda.

Groove Quantum was not asked for comment for this article. The reporting is based on publicly available sources, listed below.