List your company

Long-term Conversational Memory Question Answering on LoCoMo (test)

60.2Single Hop F1H-MEM
42.2646.917551.57556.2325May 15, 2026

Evaluation results

MethodLinks
H-MEMBackbone=GPT-4o-mini2026.0560.295.0147.589.016180.0628.463.5456.0688.83
H-MEMBackbone=GPT-4.1-mini2026.0557.6796.0846.8189.016489.7234.972.9255.5892.01
H-MEMBackbone=GPT-4.1-mini, k=302026.0556.896.3146.7993.6264.4190.0330.6269.7954.9292.86
EverMemOSBackbone=GPT-4.1-mini2026.0553.7994.2945.2290.4363.2688.1633.6669.7952.9490.78
MemOSBackbone=GPT-4.1-mini2026.0552.1687.6341.3679.7961.5877.8830.6568.7550.882.99
ZepBackbone=GPT-4.1-mini2026.0551.7688.4738.2274.1157.7875.3928.1767.7149.0681.81
Mem0Backbone=GPT-4.1-mini2026.0549.0872.5338.7760.2849.2257.6329.7857.2946.0266.23
EverMemOSBackbone=GPT-4o-mini2026.0548.7289.0639.0385.1158.7982.8728.1164.5847.7685.52
MemTreeBackbone=GPT-4.1-mini2026.0548.3578.3640.2565.2550.967.2930.158.3346.2672.4
MemoryOSBackbone=GPT-4.1-mini2026.0547.268.4939.862.0650.362.6229.451.0445.3865
MemOSBackbone=GPT-4o-mini2026.0546.9582.5236.9574.4755.2472.926.7962.545.5977.79
ZepBackbone=GPT-4o-mini2026.0546.8183.2333.5769.1555.9370.0924.2661.4644.8876.56
Mem0Backbone=GPT-4o-mini2026.0545.0167.334.6454.9644.3653.2724.8952.0841.7261.17
MemTreeBackbone=GPT-4o-mini2026.0544.972.7735.6660.6445.1162.325.4352.0842.0467.08
MemoryOSBackbone=GPT-4o-mini2026.0542.9563.8535.1856.7444.4157.6324.6345.8340.6960.13