12 December 2025

Executive Summary (TL;DR)

Dutch startup QuantWare has announced VIO‑40K, a quantum processing architecture capable of delivering 10,000 qubits, a hundredfold increase over today’s typical 100‑qubit chips quantware.com. Using a 3D scaling design, VIO‑40K links thousands of chiplets via high‑fidelity connections, providing 40,000 input‑output lines and drastically reducing the cost per qubit. The architecture integrates with NVIDIA’s NVQLink and is open to any organization working with superconducting qubits. QuantWare will build an industrial “Kilofab” facility to manufacture the chips at scale. For entrepreneurs, this jump in qubit count could accelerate the commercial viability of quantum computing across chemistry, logistics and materials science.

Article

Quantum computing has long been described as the next great leap in computation, promising to tackle problems beyond the reach of classical machines. Yet progress has been slow: after a decade of hype, most commercial quantum processors top out around 100 qubits. Google moved from 53 to 105 qubits in six years, and IBM only recently announced a 120‑qubit chip. Scaling limitations, particularly the challenge of controlling large numbers of qubits with high fidelity, have been the primary bottleneck.

Enter QuantWare, a spin‑out from TU Delft’s QuTech research institute. On 9 December 2025, the company unveiled VIO‑40K™, a new architecture designed to smash the scaling barrier. VIO‑40K is essentially a modular system of chiplets interconnected via ultra‑high‑fidelity 3D connections. These connections provide 40,000 input‑output lines, enabling tens of thousands of qubits to be controlled without the cross‑talk and errors that plague traditional designs. Because the architecture is modular, it can scale any existing superconducting qubit design up to “megaqubit” scale.

The impact is profound. QuantWare claims that a single VIO‑40K chip packs 10,000 qubits, 100 times more than any commercially available quantum processor today. Extrapolate that scaling into future devices and the quantum advantage moves from theoretical to practical. For perspective, doubling qubits exponentially increases quantum computing power; going from 100 to 10,000 qubits isn’t merely 100 × more powerful—it represents a transition from toy problems to real‑world applications.

QuantWare’s architecture also tackles cost. Current quantum roadmaps rely on linking many small chips over relatively low‑fidelity networks, creating complex systems with high operational expenses. By contrast, VIO‑40K’s 3D interconnects deliver higher performance per dollar and per watt. That means companies could build quantum accelerators that are both more powerful and less expensive to operate.

Integration is another key element. QuantWare announced that NVIDIA’s NVQLink will be part of the VIO‑40K ecosystem NVQLink provides low‑latency, high‑throughput connections between quantum and classical processors, enabling hybrid computation where quantum chips accelerate specific tasks while classical GPUs handle the rest. The combination promises to unlock “hyperscaled” quantum‑AI workflows accessible through NVIDIA’s CUDA‑Q platform.

To manufacture these chips, QuantWare plans to build Kilofab, the world’s first dedicated fab for Quantum Open Architecture devices. Scheduled to open in 2026, Kilofab will scale production capacity twentyfold, making the Netherlands a hub for commercial quantum hardware. CEO Matt Rijlaarsdam emphasized that, for years, the industry has hyped quantum’s potential without solving the scaling problem. “VIO finally removes this barrier,” he said, positioning QuantWare as a catalyst for economically relevant quantum computers.

For entrepreneurs, VIO‑40K opens new avenues. Startups in quantum chemistry could model complex molecules or catalysts; logistics companies might optimize routes beyond the reach of classical algorithms; materials scientists could design novel superconductors. The open architecture means businesses don’t need to develop proprietary qubits; they can leverage their existing designs and scale them within VIO‑40K.

However, commercialization will take time. Reservations for VIO‑40K chips open now, but deliveries aren’t expected until 2028. In the meantime, early adopters should experiment with existing cloud‑based quantum services to understand how quantum acceleration fits into their workflows. They should also monitor standards and interoperability: as multiple hardware vendors emerge, ensuring software portability across platforms will be crucial.

• Start exploring quantum use cases: Identify problems in your business that may benefit from quantum acceleration, logistics optimization, complex simulations or cryptography, and prototype them on existing cloud‑based quantum platforms.

• Plan for hybrid architectures: VIO‑40K’s integration with NVIDIA’s NVQLink means quantum will complement classical computing. Build teams that understand both worlds.

• Watch hardware roadmaps: Quantum hardware evolves quickly. Keep tabs on vendors like QuantWare to anticipate when qubit counts and error rates reach your application’s threshold.

• Engage in ecosystem conversations: As open architectures like VIO emerge, contribute to standards discussions to ensure interoperability and avoid vendor lock‑in.

QuantWare’s VIO‑40K signals that quantum computing is leaving its infancy. With a modular design that scales to 10,000 qubits and integrates seamlessly with classical systems, the architecture could usher in a new era of commercially viable quantum applications. Forward‑thinking entrepreneurs who begin experimenting now will be well positioned to harness quantum’s transformative potential when hardware arrives.

Subscribe to our newsletter to stay up to date with the latest breakthroughs in AI, Business, Technology, and Mindset.