Autonomous Code Migration for Quantum Programming Languages: A Case Study with QED-C Benchmarks
Abstract
Quantum programming requires concepts and intuition that differ substantially from those used in traditional programming languages. Multiple quantum programming frameworks address these requirements through different abstractions and execution models, creating tradeoffs in programmability, portability, and performance. Because quantum software is less prevalent in the training data available to large language models, autonomous coding agents might be expected to face additional challenges when working with these frameworks. We present a case study in autonomous migration of the QED-C Application-Oriented Benchmarks from Qiskit to CUDA-Q. Opus, GPT, and Gemini models were each tasked with porting thirteen benchmark methods spanning foundational algorithms, amplitude estimation, Monte Carlo methods, variational optimization, linear-system solving, and factoring. In all cases, the models were able to generate satisfactory ports with little direct coding help and there was a drastic reduction in the time required compared to a human engineer. However, in all cases, the agents resorted to shortcuts when implementation challenges surfaced and multiple prompts were required to force appropriate coding directions. Assessment of the generated implementations showed that no single model consistently produced the best port across all benchmarks, indicating that multi-agent ensemble strategies could mitigate challenges associated with quantum software. The effort resulted in publicly available, near-complete CUDA-Q support in the QED-C benchmarks along with model-generated skills designed to improve agent performance when implementing CUDA-Q applications. When tested on an unseen benchmark, these skills reduced the required model turns by up to 85% and token usage by up to 96%. We conclude by distilling lessons from the case study.
Related
- igraph 1.0 enables fast and robust network analysis across programming languages
- Beyond Spectroscopic Strong Coupling: Operational Benchmarks for Cavity-to-Magnon Quantum Transfer
- Neutrino Spin--Flavor Dynamics and Quantum Rabi Model: Spectral Correspondence and QRM Benchmarks
- Representation Geometry and Inductive Biases in Quantum Classifiers across Classical Visual Benchmarks
- Attainable Boundaries versus Operational Equivalence: Dual Benchmarks for the Hierarchical Assessment of Quantum Resources
BibTeX
@misc{brown2026autonomous,
title = {{Autonomous Code Migration for Quantum Programming Languages: A Case Study with QED-C Benchmarks}},
author = {W. Michael Brown and Thien Nguyen and Avimita Chatterjee and Thomas Lubinski},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.12187},
eprint = {2610.12187},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.12187},
}
RIS
TY - UNPB
TI - Autonomous Code Migration for Quantum Programming Languages: A Case Study with QED-C Benchmarks
AU - W. Michael Brown
AU - Thien Nguyen
AU - Avimita Chatterjee
AU - Thomas Lubinski
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.12187
UR - https://arxiv.org/abs/2610.12187
AB - Quantum programming requires concepts and intuition that differ substantially from those used in traditional programming languages. Multiple quantum programming frameworks address these requirements through different abstractions and execution models, creating tradeoffs in programmability, portability, and performance. Because quantum software is less prevalent in the training data available to large language models, autonomous coding agents might be expected to face additional challenges when working with these frameworks. We present a case study in autonomous migration of the QED-C Application-Oriented Benchmarks from Qiskit to CUDA-Q. Opus, GPT, and Gemini models were each tasked with porting thirteen benchmark methods spanning foundational algorithms, amplitude estimation, Monte Carlo methods, variational optimization, linear-system solving, and factoring. In all cases, the models were able to generate satisfactory ports with little direct coding help and there was a drastic reduction in the time required compared to a human engineer. However, in all cases, the agents resorted to shortcuts when implementation challenges surfaced and multiple prompts were required to force appropriate coding directions. Assessment of the generated implementations showed that no single model consistently produced the best port across all benchmarks, indicating that multi-agent ensemble strategies could mitigate challenges associated with quantum software. The effort resulted in publicly available, near-complete CUDA-Q support in the QED-C benchmarks along with model-generated skills designed to improve agent performance when implementing CUDA-Q applications. When tested on an unseen benchmark, these skills reduced the required model turns by up to 85% and token usage by up to 96%. We conclude by distilling lessons from the case study.
ER -