FirstQFM reports achieving 33% shorter gate times on a QuantWare Contralto-D21 quantum processor by applying its quantum foundation model, validated through randomized benchmarking without loss of gate fidelity. The model generated optimized pulse-level gates using calibration data from Quantum Machines, shifting optimization from quantum hardware to classical compute.
Researchers also identified improvements to five qubit pairs, demonstrating higher Bell-state fidelity confirmed with independent hardware runs, a key step in tackling the challenges of entanglement and two-qubit calibration. “FirstQFM’s proprietary, patented foundation models are built to learn the operating characteristics of quantum hardware and use that information to improve how the system performs,” said Vish Ramakrishnan, CEO and Co-Founder of FirstQFM.
FirstQFM Model Achieves 33% Faster Single-Qubit Gates on QuantWare QPU
By shifting optimization to classical computing resources, FirstQFM reduces demand on limited and expensive quantum processing unit time, a critical step toward scaling quantum computations. The model’s impact extends beyond single-qubit operations, with improvements also identified in two-qubit gate performance, a traditionally challenging area for calibration and essential for creating entanglement. FirstQFM’s approach pinpointed five qubit pairs exhibiting higher Bell-state fidelity, with each enhancement confirmed through independent hardware runs at the Israeli Quantum Computing Center.
The foundation for these gains lies in the model’s design, which uses NVIDIA accelerated computing hardware and software, including the Leonardo supercomputer and the CUDA-Q platform. FirstQFM developed its foundation model using these resources before applying and fine-tuning it to the QuantWare processor.
This integration of classical and quantum computing infrastructure allowed the team to efficiently analyze calibration data and generate optimized gate implementations, the company says. As superconducting quantum processors increase in complexity, calibration presents a growing control problem, with more qubits and couplers introducing greater operating parameters and potential for device interactions. FirstQFM intends to expand the range of optimizations available for the 21-qubit device and scale the solution in anticipation of QuantWare’s advancements toward larger, VIO-based quantum processing units.
On QuantWare’s processor, our model was able to use existing calibration data to generate shorter single-qubit gates while preserving fidelity, and to improve selected two-qubit operations.
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
