Quantum circuits help AI overcome memory limitations with minimal new parameters

For millions of people, chatbots powered by large language models are now a key feature of everyday life. These AI systems are growing at a rapid pace, but scaling them up is becoming increasingly costly and resource-intensive. The science-journalism coverage adds useful context, while the strongest evidential footing still comes from the underlying data, papers or institutional documentation.

That matters because physics only takes a result seriously when the measurement chain remains robust under scrutiny. Experimental particle physics and precision metrology both operate in regimes where the signal sits far below the background noise, and where systematic uncertainties can mimic new physics if not controlled rigorously. The history of the field contains numerous anomalies that generated theoretical excitement before better data showed them to be artifacts, and it also contains genuine discoveries that were initially dismissed as noise. The difference is almost always resolved by independent replication with different instruments and different systematics. This article has been reviewed according to Science X's editorial process and policies. Through a new preprint on the arXiv server, a team led by Borja Aizpurua at Multiverse Computing in San Sebastián, Spain, has found a way to improve the performance of LLMs using.

GPT-5.5, for instance, is estimated to contain somewhere between two and five trillion parameters. Rather than adding vast numbers of new classical parameters, they inserted small quantum circuit blocks into the inner workings of a pre-trained LLM.

The resulting system is a hybrid: the original LLM runs on a standard computer, while the quantum components are executed on IBM's 156-qubit superconducting quantum processor. The team also tested their platform on SmolLM2, a smaller 135-million-parameter model chosen because it was more tractable to study systematically.

But in demonstrating that quantum enhancement can work at all on a real, widely used model, their results are already promising. We rely on readers like you to keep independent science journalism alive.

The broader interest lies as much in the method as in the headline number, because a durable measurement procedure can travel farther than a single result. When experimental physicists develop a technique that achieves new sensitivity or controls a previously uncharacterized systematic, that methodological contribution persists even if the specific measurement is later revised. This is one reason why precision physics experiments often generate long-term value that is not immediately visible in the original publication.

Borja Aizpurua et al, Quantum-enhanced Large Language Models on Quantum Hardware via Cayley Unitary Adapters, arXiv (2026). Full profile → MA in English, copy editor since 2021 with experience in higher education and health content.

Because this item comes through Phys. org Physics as science journalism, it should be treated as contextual reporting rather than primary evidence. Good science reporting can identify why a result matters, connect it to the wider literature and make technical work readable, but the decisive evidence remains in the original paper, dataset, mission release or technical record. That distinction is especially important when a story is later repeated by aggregators, because repetition increases visibility, not evidential strength.

The next step is more measurement, tighter systematic control and scrutiny from groups whose experimental setups are genuinely independent. In experimental particle physics and precision metrology, the threshold for a discovery claim is a five-sigma excess surviving multiple analyses; an intriguing signal at lower significance is a reason to run more experiments, not a reason to revise the textbooks. Next-generation experiments currently under construction or commissioning will revisit several of the open questions that give the current result its context.

Source