ResearchBenchmarksStrategic AI Persuasion (Receiver) on MixTalk Large EnvironmentFollow1.78BTgem-3fpr0.00160.46330.9251.3867Feb 1, 2026Evaluation ResultsMethodMethodLinksBTMean ScoreTORgem-3fpr2026.021.780.570.14gpt-5m2026.021.440.560.16grok-4.1f2026.021.020.50.23Kimi-k2t2026.020.690.40.31gpt-4.1m2026.020.070.020.74