BenchmarkHub: an open-source platform for reproducible evaluation of large language models in healthcare
Background: Large language models (LLMs) are solely judged by their scores on biomedical benchmarks, yet these scores are rarely reproducible since the pro