Fujitsu Unveils a Diamond-Spin Quantum Computer Prototype

Fujitsu’s September 8, 2026 announcement was a hardware milestone, not the launch of a large-scale fault-tolerant computer. The company said it had developed the world’s first working prototype of a diamond-spin quantum computer with tin-vacancy (SnV) centers integrated directly into photonic integrated circuits. The system is quantum-gate operable and built as a foundation for linking future modules optically. 211

What Fujitsu built

The prototype uses SnV centers in diamond as spin qubits. Fujitsu integrated those qubits with on-chip photonic circuitry, allowing light to be used in qubit control and readout and, in the company’s proposed scaling model, to connect quantum-computing modules. 211

This is a distinct hardware route from superconducting processors. Fujitsu’s emphasis is not simply on replacing superconducting qubits in the near term, but on developing a platform that could combine high-fidelity diamond-spin operations with optical interconnects for modular expansion. 2

Why the photonic integration matters

A major challenge in scaling quantum hardware is connecting useful qubits without losing the fragile quantum information they hold. Fujitsu’s prototype places the diamond qubits and photonic components in the same system so emitted and control light can be routed through optical waveguides. Its technical materials describe alumina optical-waveguide circuits designed for the visible wavelengths used in SnV control and signal light. 11

Fujitsu attributes the result to three enabling areas:

• quantum-circuit conversion and qubit-control technology for the diamond-spin hardware;
• heterogeneous bonding and thinning of diamond with alumina/silicon-dioxide structures; and
• photonic-integrated-circuit fabrication compatible with SnV centers. 213

Together, those advances are intended to make the physical qubit module usable through a conventional quantum-computing workflow rather than only as a laboratory device.

Temperature and platform access

Fujitsu said the prototype operates at approximately −271.6°C. That is warmer than the approximately −273.13°C operating temperature cited for typical superconducting quantum computers, but it remains a cryogenic system and does not eliminate the need for specialized cooling infrastructure. 2

The company also demonstrated use of the prototype through its Fujitsu Hybrid Quantum Computing Platform. Fujitsu says its circuit-conversion capability lets users submit quantum circuits without needing to work directly with the particulars of the diamond-spin hardware. 213

Research behind the prototype

The work follows a research collaboration that began in 2020 among Fujitsu, Delft University of Technology and QuTech. 12

Fujitsu also points to prior diamond-spin research, including a March 2025 demonstration of a universal gate set with an error probability below 0.1%, as well as earlier remote nitrogen-vacancy-center entanglement and gate-operation demonstrations. These results form part of the research basis for the new SnV-and-photonics prototype. 2

What comes next—and what the prototype is not

The newly announced machine is a single-module prototype. It should not be confused with a 250-logical-qubit or 1,000-logical-qubit computer. Those figures are Fujitsu development targets, not demonstrated capacities of this device. 28

Fujitsu’s stated roadmap includes:

• a multi-module diamond-spin prototype in 2027;
• further work to integrate diamond-spin and superconducting hardware after 2030;
• a target of 250 logical qubits in fiscal 2030; and
• a target of 1,000 logical qubits in fiscal 2035. 238

Those plans sit alongside, rather than replace, Fujitsu’s superconducting program. In 2025, the company began R&D toward a superconducting machine with more than 10,000 physical qubits, targeted for completion in fiscal 2030 and designed to operate with 250 logical qubits. 36

The broader significance

The announcement signals that Fujitsu is pursuing more than one path to scalable quantum computing: superconducting hardware for a large physical-qubit system and diamond-spin modules for optical interconnection and longer-term hybrid architectures. Whether the diamond-spin approach can scale beyond a prototype will depend on the next steps—especially demonstrating reliable optical links between modules—rather than on the single-module milestone alone. 28

The timing also highlights a changing Japanese quantum-hardware landscape. Reports said NEC ended physical quantum-computer hardware development at the end of March 2026 amid commercialization challenges, while Fujitsu continues to invest in both superconducting and diamond-spin approaches. The NEC development was reported by news organizations rather than announced as a product release by NEC. 1718