Is one run enough? Reproducibility of flagship large language models across temperature and reasoning settings in biomedical text processing

For binary biomedical classification with tightly constrained outputs, both models were reproducible across decoding and reasoning settings, suggesting single runs are often sufficient, with minimal replication as a practical stability check.