ResearchBenchmarksMathematical Reasoning on Generalized GSM8K VeRA (test)Follow98.56Pass@1 AccuracyGemini 2.5 Pro93.63296.09698.56101.024Jan 23, 2026Evaluation ResultsMethodMethodLinksPass@1 AccuracyGemini 2.5 Pro2026.0198.56OpenAI GPT-52026.0198.56