394 hidden bugs found in leading quantum simulators—most aren't quantum flaws

A new study has uncovered 394 confirmed bugs in 12 open-source quantum simulators. Researchers carefully filtered the data to focus only on verified failures, excluding feature requests or documentation updates. The findings reveal that many issues originate not from quantum logic but from classical infrastructure flaws.

The dataset of 394 bugs was compiled by analysing merged pull requests, ensuring only confirmed failures were included. Of these, around 40% were quantum-specific, while the rest stemmed from classical components like memory management. In fact, 63.2 bugs were directly linked to classical infrastructure problems rather than quantum execution errors.

Silent logical errors—where programs produce seemingly correct but wrong results—made up roughly 100 of the total bugs. Many of these went undetected by automated testing, highlighting gaps in current verification methods. The study also noted that as quantum programs grow more complex, the reliability of classical components becomes increasingly critical. Researchers emphasised that better classical software engineering practices could significantly reduce failures in quantum simulations. While the study did not name specific platforms or authors, it underscored the broader challenge of ensuring stability in quantum software development.

The findings point to a clear need for stronger classical infrastructure in quantum simulators. With 394 confirmed bugs identified, the study suggests that improving traditional software practices could enhance overall reliability. Silent errors and undetected logical flaws remain key concerns as the field advances.