FirstQFM cuts gate times by 33% on QuantWare QPU

FirstQFM says its quantum foundation model reduced single-qubit gate execution times by 33% on a QuantWare superconducting processor at the Israeli Quantum Computing Center, while preserving fidelity. The results suggest more of the optimization burden can move from scarce QPU time to classical compute as quantum hardware scales.

Why it matters: - FirstQFM’s approach aims to speed up quantum hardware tuning without consuming extra quantum processor time. - The results point to a way to improve gate performance on superconducting QPUs as calibration complexity rises with more qubits and couplers. - The work could help make quantum systems more useful, scalable and reliable by shifting more optimization onto classical computing.

What happened: - FirstQFM announced results from applying its quantum foundation model to a QuantWare superconducting quantum processor at the Israeli Quantum Computing Center. - The model generated single-qubit gates with 33% shorter execution times while preserving gate fidelity. - The results were validated on a QuantWare Contralto-D21 QPU using randomized benchmarking. - The model also identified improvements to two-qubit gates, which are central to entanglement and harder to calibrate than single-qubit operations.

The details: - FirstQFM generated custom pulse-level gate implementations from calibration data produced by Quantum Machines’ calibration stack. - The single-qubit gate improvements required no additional quantum hardware time. - Using existing calibration data plus a short hardware probe, the model identified five qubit pairs with higher Bell-state fidelity. - Each two-qubit improvement was confirmed in independent hardware runs. - The QuantWare processor at IQCC ran with Quantum Machines’ OPX control hardware, QUA software and automated calibration infrastructure. - FirstQFM developed the foundation model using NVIDIA accelerated computing hardware and software, including the Leonardo supercomputer and the NVIDIA CUDA-Q open source hybrid platform, before fine-tuning and applying the model to the QuantWare processor. - FirstQFM said its proprietary, patented foundation models are built to learn the operating characteristics of quantum hardware and use that information to improve system performance. - FirstQFM said the model used existing calibration data to generate shorter single-qubit gates while preserving fidelity and to improve selected two-qubit operations. - The company’s core technology is protected by allowed and pending patents. - FirstQFM is based in Stockholm. - QuantWare is headquartered in Delft, the Netherlands. - QuantWare says it has shipped more quantum processors than any other commercial supplier and serves more than 50 customers across 20 countries.

Between the lines: - The technical milestone is less about one faster gate and more about proving that foundation models can mine calibration data for practical control gains. - The result matters because quantum hardware time is scarce and expensive, so moving optimization into classical compute could improve development efficiency. - The focus on two-qubit gates is important because entanglement operations tend to be the bottleneck for useful quantum computation. - The collaboration also ties together hardware, control software and accelerated classical compute, which is a sign that quantum progress increasingly depends on the full stack rather than on qubits alone.

What’s next: - FirstQFM plans to expand the set of optimizations available for the 21-qubit device. - The company also plans to scale the solution as QuantWare moves toward larger VIO-based QPUs. - Further validation on larger devices would show whether the model’s gains hold as calibration problems grow more complex.

The bottom line: - FirstQFM is pitching foundation models as a practical tool for quantum control, with early hardware-validated gains on a QuantWare QPU.