OpenProblemBench is a benchmark comprising 82 unresolved problems drawn from mathematics and theoretical physics literature. Each problem supplies the research context, assumptions, and prior progress, and is chosen so that proposed solutions can be checked by clear mathematical or computational criteria.\ \ Four evaluator models independently assess the correctness, completeness, and degree of progress of each submission without reference solutions. Across seven evaluated configurations, GPT-6-Astra achieves the highest mean judged solve rate of 14.0%, compared with 5.5%–6.7% for full-size open models and 2.4%–3.7% for Flash models.\ \ Case comparisons link stronger outcomes to changes in problem representation, the use of general arguments that extend beyond finite evidence, and proofs of the steps needed to complete a solution. By grounding evaluation in questions arising from the research literature, OpenProblemBench provides a setting for investigating AI's capabilities and limitations as a contributor to foundational theoretical science.\ \ Review