OpenAI has introduced GeneBench-Pro, a benchmark suite focused on genomics-related tasks. The release is part of a broader effort to evaluate how AI systems perform in specialized scientific domains.

Domain benchmarks matter because general reasoning scores do not necessarily show whether a model can handle biological data, terminology or research workflows. Genomics tasks require accuracy, context and careful uncertainty handling.

For scientific AI, GeneBench-Pro gives researchers another tool for comparing systems before they are used in sensitive or high-cost work.