We design and implement a quantum combinatorial reasoning framework for large language models (QCR-LLM), integrating a real quantum computer in the hybrid workflow. QCR-LLM reformulates reasoning aggregation as a higher-order unconstrained binary optimization problem. In this sense, reasoning fragments are represented as binary variables and their interactions encode statistical relevance, logical coherence, and semantic redundancy. We tackle the resulting high-order optimization problem both classically, via simulated annealing, and quantumly, through the bias-field digitized counterdiabatic quantum optimizer executed on IBM's superconducting digital quantum processors. Experiments on BIG-Bench Extra Hard benchmarks demonstrate that our QCR-LLM consistently improves reasoning accuracy across multiple LLM backbones, surpassing reasoning-native systems such as o3-high and DeepSeek R1 by up to $+9$ pp. At a matched sampling budget, QCR-LLM also outperforms standard chain-of-thought aggregation baselines (majority vote and self-consistency), as well as a pairwise Quadratic Unconstrained Binary Optimization restriction of our formulation, indicating that the improvement arises from the higher-order energy landscape rather than from multisampling alone. Despite requiring multiple reasoning samples per query, our QCR-LLM reduces the number of generated tokens and the associated LLM-side energy relative to long hidden chain-of-thought models such as o3-high; we emphasize, however, that once the energy of the quantum backend is accounted for, we are not yet in an energy-saving or sustainability regime. These results constitute the first experimental evidence of quantum-assisted reasoning, showing that hybrid quantum-classical optimization can efficiently enhance reasoning coherence and interpretability in large-scale language models. We have opened the doors to the emergence of quantum intelligence, where the higher-order structure of harder prompts maps naturally onto quantum optimizers, opening a paradigm toward genuine quantum reasoning; while we observe a small accuracy gain of the quantum solver over simulated annealing, this result is not conclusive and may stem from statistical fluctuations.

Close

PrevNext

Reuse & PermissionsView figure in article

• 
• 

• Artificial intelligence
• Natural language processing
• Quantum algorithms
• Quantum algorithms & computation
• Quantum benchmarking
• Quantum circuits
• Quantum computation