SQC Accelerates Atomic-Scale Chip Iteration with NVIDIA Ising Open Models

Today at NVIDIA GTC Berlin, an inside look at how we’re using NVIDIA Ising Calibration to reduce QPU screening time and accelerate our design-build-test loop.

‍

Silicon Quantum Computing (SQC) is the global pioneer of atomic-scale semiconductor manufacturing, building QPUs by placing individual phosphorus atoms in pure silicon with 0.13 nanometer precision – 100x smaller than the world’s most advanced classical processes.

Core to this work is our one-week prototyping cycle: manufacturing and measuring new devices to inform sub-nanometer design improvements. Measurement is similar to the test-and-yield approach used in traditional semiconductor manufacturing, where each chip informs the next. Rapid, low-noise measurement is a key part of our iterative process and, accordingly, SQC uses the best hardware and software tools on the market.

This design-build-test-iterate loop is a critical process that feeds directly into the next generation of our atomic-scale chips. The more we automate and accelerate it, the faster we scale toward the millions of qubits required for our universal, fault-tolerant quantum computer.

That’s where NVIDIA Ising comes in.

Using NVIDIA Ising Calibration, part of the NVIDIA Ising open model family for quantum computing, we have significantly reduced the time it takes to screen our QPUs. A process that previously took three hours has been reduced to one, with the greatest benefit coming from highly customizable, open agentic workflows enabling faster parameter extraction across devices. Device data is captured in a live database and fed directly to our atomic fabrication team, creating a streamlined feedback loop.

Put simply, NVIDIA Ising enables a massive step forward in productivity: freeing up engineers’ time and recycling it into our overall process helps SQC move faster.

We are now pushing this automation even further. The next step is to run Ising Calibration continuously, significantly reducing the downtime required for scheduled device calibrations, or tunings. We are also deploying NVIDIA Ising across multiple projects to recycle time into core work. At industrial scale, Ising will allow SQC to seamlessly calibrate hundreds of devices in parallel: an essential capability for commercial production.

Our expanding integration with the NVIDIA quantum platform.

In October 2025, SQC announced work with NVIDIA NVQLink upon its launch: high-speed interconnects enabling real-time orchestration of QPUs alongside CPUs and GPUs.

Where NVIDIA NVQLink provides a reference architecture for hybrid quantum-classical integrations, NVIDIA Ising accelerates our speed of atomic-scale chip iteration. Together, these efforts mark a deepening collaboration with NVIDIA focused on delivering commercially useful, fault-tolerant quantum computing.

“Accelerating our commercial roadmap requires a global network of partners that can match the precision and performance that SQC’s qubits demand,” said Michelle Simmons, Founder and CEO of SQC. “NVIDIA is exactly that kind of partner, and we look forward to expanding our work together as we build toward universal, fault-tolerant quantum systems.”