ResearchBenchmarksGeneral AI Assistant Tasks on GAIA Level 2 (val)Follow54.2AccuracyGPT-521.9630.3338.747.07Dec 7, 2025Evaluation ResultsMethodMethodLinksAccuracy95% Confidence IntervalBetween-group VarianceIntraclass Correlation Coefficient (ICC)GPT-5Model=GPT-5, Web searc...Model=GPT-5, Web search=true, Trials per question=642025.1254.2—0.1870.745GPT-4oModel=GPT-4o, Web sear...Model=GPT-4o, Web search=true, Trials per question=642025.1223.2—0.1190.662