QuanBench+: A Unified Multi-Framework Benchmark for LLM-Based Quantum Code Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Slim, Ali, Hamieh, Haydar, Kotaich, Jawad, Ghosn, Yehya, Chehimi, Mahdi, Mohanna, Ammar, Hammoud, Hasan Abed Al Kader, Ghanem, Bernard
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914498721349632
author Slim, Ali
Hamieh, Haydar
Kotaich, Jawad
Ghosn, Yehya
Chehimi, Mahdi
Mohanna, Ammar
Hammoud, Hasan Abed Al Kader
Ghanem, Bernard
author_facet Slim, Ali
Hamieh, Haydar
Kotaich, Jawad
Ghosn, Yehya
Chehimi, Mahdi
Mohanna, Ammar
Hammoud, Hasan Abed Al Kader
Ghanem, Bernard
contents Large Language Models (LLMs) are increasingly used for code generation, yet quantum code generation is still evaluated mostly within single frameworks, making it difficult to separate quantum reasoning from framework familiarity. We introduce QuanBench+, a unified benchmark spanning Qiskit, PennyLane, and Cirq, with 42 aligned tasks covering quantum algorithms, gate decomposition, and state preparation. We evaluate models with executable functional tests, report Pass@1 and Pass@5, and use KL-divergence-based acceptance for probabilistic outputs. We additionally study Pass@1 after feedback-based repair, where a model may revise code after a runtime error or wrong answer. Across frameworks, the strongest one-shot scores reach 59.5% in Qiskit, 54.8% in Cirq, and 42.9% in PennyLane; with feedback-based repair, the best scores rise to 83.3%, 76.2%, and 66.7%, respectively. These results show clear progress, but also that reliable multi-framework quantum code generation remains unsolved and still depends strongly on framework-specific knowledge.
format Preprint
id arxiv_https___arxiv_org_abs_2604_08570
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle QuanBench+: A Unified Multi-Framework Benchmark for LLM-Based Quantum Code Generation
Slim, Ali
Hamieh, Haydar
Kotaich, Jawad
Ghosn, Yehya
Chehimi, Mahdi
Mohanna, Ammar
Hammoud, Hasan Abed Al Kader
Ghanem, Bernard
Machine Learning
Artificial Intelligence
Programming Languages
Software Engineering
Quantum Physics
Large Language Models (LLMs) are increasingly used for code generation, yet quantum code generation is still evaluated mostly within single frameworks, making it difficult to separate quantum reasoning from framework familiarity. We introduce QuanBench+, a unified benchmark spanning Qiskit, PennyLane, and Cirq, with 42 aligned tasks covering quantum algorithms, gate decomposition, and state preparation. We evaluate models with executable functional tests, report Pass@1 and Pass@5, and use KL-divergence-based acceptance for probabilistic outputs. We additionally study Pass@1 after feedback-based repair, where a model may revise code after a runtime error or wrong answer. Across frameworks, the strongest one-shot scores reach 59.5% in Qiskit, 54.8% in Cirq, and 42.9% in PennyLane; with feedback-based repair, the best scores rise to 83.3%, 76.2%, and 66.7%, respectively. These results show clear progress, but also that reliable multi-framework quantum code generation remains unsolved and still depends strongly on framework-specific knowledge.
title QuanBench+: A Unified Multi-Framework Benchmark for LLM-Based Quantum Code Generation
topic Machine Learning
Artificial Intelligence
Programming Languages
Software Engineering
Quantum Physics
url https://arxiv.org/abs/2604.08570