ResearchBenchmarksOverall reasoning performance on LoCoMo (test)Follow80.6LLM ScoreFullContext30.99243.87156.7569.629Jan 13, 2026Evaluation ResultsMethodMethodLinksLLM ScoreF1 ScoreBLEU-1FullContextLLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0180.653.345NemoriLLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0179.251.944.5LangMemLLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0173.447.640SwiftMemLLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0170.442.946.7Mem0LLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0166.343.536.5ZepLLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0161.636.930.9RAGLLM=GPT-4.1-miniLLM=GPT-4.1-mini2026.0132.923.519.2