ResearchBenchmarksLong-term Conversational Memory on LoCoMo Open-domainFollow12.44B1 ScoreSAGE10.921611.315811.7112.1042May 29, 2026Evaluation ResultsMethodMethodLinksB1 ScoreF1 ScoreJ ScoreSAGEAveraged across=7 open...Averaged across=7 open-weight models, LLM Judge=Llama-3.1-8b2026.0512.4416.8788.09Mem0Averaged across=7 open...Averaged across=7 open-weight models, LLM Judge=Llama-3.1-8b2026.0511.3415.0482.79Mem0^gAveraged across=7 open...Averaged across=7 open-weight models, LLM Judge=Llama-3.1-8b2026.0510.9814.6681.89