ResearchBenchmarksMemory-intensive task on LongMemEvalMemory-intensive task on LongMemEvalFollow57.41F1 ScoreNapMem21.290830.667940.04549.4221Jul 7, 2026Evaluation resultsMethodMethodLinksF1 ScoreLLM-Judge Accuracy (L-J)NapMemModel Size=9B, Optimiz...Model Size=9B, Optimization=w/ RL2026.0757.4180.33Mem02026.0753.8678NapMemModel Size=397BModel Size=397B2026.0753.8578.33Zep2026.0752.6173.33NapMemModel Size=122BModel Size=122B2026.0748.8670MemOS2026.0739.1654.33AgeMem2026.0737.8351.33MemoryOS2026.0722.6823.67