ResearchBenchmarksLong-horizon Dialogue Reasoning on LoCoMo (test)Follow44.65F1 ScoreAdaMem25.638830.574435.5140.4456Mar 17, 2026Evaluation ResultsMethodMethodLinksF1 ScoreBLEU-1 ScoreAdaMemTokens=2248, Latency (...Tokens=2248, Latency (s)=4.7222026.0344.6537.92Mem0Tokens=1340, Latency (...Tokens=1340, Latency (s)=3.7392026.0338.1632.04ZepTokens=2461, Latency (...Tokens=2461, Latency (s)=3.2552026.0332.427.09A-MemTokens=2720, Latency (...Tokens=2720, Latency (s)=3.2272026.0326.3720.7