ResearchBenchmarksLong-term Conversational Memory on LoCoMo Single-hopFollow9.28B1 ScoreSAGE8.48968.69488.99.1052May 29, 2026Evaluation ResultsMethodMethodLinksB1 ScoreF1 ScoreJ ScoreSAGEAveraged across=7 open...Averaged across=7 open-weight models, LLM Judge=Llama-3.1-8b2026.059.2813.685.11Mem0Averaged across=7 open...Averaged across=7 open-weight models, LLM Judge=Llama-3.1-8b2026.058.7412.5479.28Mem0^gAveraged across=7 open...Averaged across=7 open-weight models, LLM Judge=Llama-3.1-8b2026.058.5212.2578.22