Long-term memory evaluation on LoCoMo
93Overall F1True Memory Pro
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| True Memory Proevaluation_protocol=3-run semantic-judge mean2026.05 | 93 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EverMemOS2026.03 | 92.3 | — | 96.1 | — | — | 91.1 | — | — | 89.7 | — | — | 70.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Hindsight Gemini-3category_mapping=mapped to LoCoMo categories, judge_protocol=custom/different2026.05 | 89.61 | — | 86.17 | — | — | 70.83 | — | — | 83.8 | — | — | 95.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Hindsight2026.03 | 89.6 | — | 86.2 | — | — | 70.8 | — | — | 95.1 | — | — | 83.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Zep v32026.03 | 85.2 | — | 90.8 | — | — | 81.9 | — | — | 77.3 | — | — | 75 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemOS v22026.03 | 80.8 | — | 85.4 | — | — | 79.4 | — | — | 75.1 | — | — | 64.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemOS2026.05 | 75.8 | — | 81.09 | — | — | 67.49 | — | — | 75.18 | — | — | 55.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SLM-V3Mode=Mode A (Retrieval)2026.03 | 74.8 | — | 72 | — | — | 70.3 | — | — | 80 | — | — | 85 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Governed Memory2026.03 | 74.8 | — | 78.7 | — | — | 51.7 | — | — | 64.6 | — | — | 83.6 | — | — | — | — | — | — | — | — | 16.4 | 34.1 | 28 | 8.2 | 13.1 | — | — | — | — | — | — | |
| memobase2026.05 | 72.01 | — | 73.12 | — | — | 64.65 | — | — | 81.2 | — | — | 53.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem02026.05 | 64.57 | — | 73.33 | — | — | 58.75 | — | — | 52.34 | — | — | 45.83 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MIRIX2026.05 | 64.33 | — | 68.22 | — | — | 54.26 | — | — | 68.54 | — | — | 46.88 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem02026.03 | 64.2 | — | 69 | — | — | 61.7 | — | — | 58.3 | — | — | 50 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OMNIMEMBackbone=GPT-5.12026.04 | 61.3 | — | 36.7 | — | — | 59.8 | — | — | 30.7 | — | — | 67.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | 74.7 | — | — | — | — | — | |
| SLM-V3Mode=Mode A (Raw)2026.03 | 60.4 | — | 57.5 | — | — | 43 | — | — | 63 | — | — | 72 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OMNIMEMBackbone=GPT-4o2026.04 | 59.8 | — | 36.5 | — | — | 55.6 | — | — | 25.5 | — | — | 64.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | 83.5 | — | — | — | — | — | |
| Zep2026.05 | 59.22 | — | 66.23 | — | — | 52.12 | — | — | 54.82 | — | — | 33.33 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AMABackbone=GPT-4o2026.01 | 58 | 0.888 | 63.6 | 57 | 0.716 | 45.3 | 32.4 | 0.764 | 60.8 | 48.4 | 0.514 | 33.3 | 25.1 | 0.805 | 49.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AMABackbone=Qwen2-72B-Instruct2026.01 | 57 | 0.872 | 62.5 | 56 | 0.703 | 44.5 | 31.8 | 0.751 | 59.7 | 47.5 | 0.505 | 32.7 | 24.7 | 0.791 | 48.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemU2026.05 | 56.55 | — | 66.34 | — | — | 63.12 | — | — | 27.1 | — | — | 50 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| AMABackbone=GPT-4o-mini2026.01 | 55.8 | 0.849 | 62.2 | 54.8 | 0.681 | 42.3 | 30.3 | 0.748 | 58.9 | 46.1 | 0.479 | 28.3 | 22.8 | 0.774 | 46.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Supermemory2026.05 | 55.34 | — | 67.3 | — | — | 51.12 | — | — | 31.77 | — | — | 42.67 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OMNIMEMBackbone=GPT-5-nano2026.04 | 52.2 | — | 37.1 | — | — | 35.7 | — | — | 25.3 | — | — | 56.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | 71.9 | — | — | — | — | — | |
| NemoriBackbone=GPT-4o2026.01 | 52.1 | 0.828 | 57.3 | 50.2 | 0.732 | 40.7 | 31.1 | 0.757 | 56.2 | 49 | 0.498 | 25.2 | 18.8 | 0.774 | 44.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FullContextBackbone=GPT-4o2026.01 | 51.9 | 0.848 | 59.9 | 52.1 | 0.753 | 43.1 | 32.9 | 0.723 | 46.3 | 39 | 0.552 | 27.7 | 21.7 | 0.786 | 43.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OMNIMEMBackbone=GPT-4o-mini2026.04 | 51.9 | — | 19.6 | — | — | 54.4 | — | — | 17.7 | — | — | 58.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | 77.9 | — | — | — | — | — | |
| FluxMem2026.02 | 51.16 | — | 62.12 | 53.52 | — | 48.56 | 39.9 | — | 56.67 | 41.88 | — | 37.3 | 29.63 | — | 41.73 | 44.05 | 54.89 | 36.76 | 62.33 | 49.51 | — | — | — | — | — | — | — | — | — | — | — | |
| AMABackbone=Qwen2-7B-Instruct2026.01 | 51 | 0.78 | 55.9 | 50 | 0.628 | 39.8 | 28.4 | 0.671 | 53.3 | 42.4 | 0.482 | 29.2 | 22.1 | 0.707 | 43.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NemoriBackbone=Qwen2-72B-Instruct2026.01 | 50 | 0.839 | 54.3 | 43.9 | 0.652 | 38.1 | 26.9 | 0.704 | 57.1 | 47.7 | 0.51 | 23.9 | 16.3 | 0.756 | 39.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NemoriBackbone=GPT-4o-mini2026.01 | 49.2 | 0.817 | 54.1 | 43 | 0.65 | 36.3 | 25.5 | 0.707 | 56.4 | 46.3 | 0.446 | 20.7 | 15 | 0.74 | 38.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| OMNIMEMBackbone=GPT-4.1-nano2026.04 | 49.2 | — | 21.6 | — | — | 47.7 | — | — | 24.4 | — | — | 58.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | 72.2 | — | — | — | — | — | |
| FullContextBackbone=Qwen2-72B-Instruct2026.01 | 46.6 | 0.845 | 52.8 | 45.3 | 0.665 | 36.9 | 27.3 | 0.555 | 44.3 | 36.9 | 0.551 | 27.3 | 18.5 | 0.733 | 39 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LangMemBackbone=GPT-4o2026.01 | 46.4 | 0.824 | 49.8 | 42.5 | 0.693 | 40.5 | 31.7 | 0.496 | 47.3 | 39.9 | 0.575 | 32 | 25.7 | 0.717 | 39 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| O-mem2026.02 | 46.4 | — | 54.89 | 48.98 | — | 42.64 | 34.08 | — | 57.48 | 49.76 | — | 30.58 | 25.69 | — | 39.63 | 32.78 | 48.88 | 21.11 | 44.26 | 36.76 | — | — | — | — | — | — | — | — | — | — | — | |
| FullContextBackbone=GPT-4o-mini2026.01 | 45.8 | 0.823 | 52.6 | 44.3 | 0.663 | 35.1 | 25.9 | 0.557 | 43.7 | 35.8 | 0.482 | 24.3 | 17.1 | 0.717 | 37.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HMOBackbone=GPT-4o-mini2026.04 | 45.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ES-MemLLM=GPT-4o-mini2026.01 | 45.56 | — | 50.07 | 45.21 | — | 36.52 | 27.75 | — | 47.9 | 42.12 | — | 24.77 | 18.51 | — | 39.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ES-MemTokens=2925, Latency (s)=1.4232026.01 | 45.56 | — | — | — | — | — | — | — | — | — | — | — | — | — | 39.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NemoriBackbone=Qwen2-7B-Instruct2026.01 | 45.4 | 0.762 | 49.3 | 39.9 | 0.592 | 34.6 | 24.4 | 0.639 | 51.9 | 43.3 | 0.463 | 21.7 | 14.8 | 0.686 | 36.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem0LLM=GPT-4o-mini2026.01 | 45.09 | — | 47.65 | 38.72 | — | 38.72 | 27.13 | — | 48.93 | 40.51 | — | 28.64 | 21.58 | — | 35.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem0Tokens=1764, Latency (s)=0.7082026.01 | 45.09 | — | — | — | — | — | — | — | — | — | — | — | — | — | 35.91 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LightMemLLM=GPT-4o-mini2026.01 | 44.73 | — | 47.64 | 40.91 | — | 32.11 | 23.79 | — | 53.79 | 46.96 | — | 26.14 | 20.19 | — | 37.74 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LightMemTokens=815, Latency (s)=1.8382026.01 | 44.73 | — | — | — | — | — | — | — | — | — | — | — | — | — | 37.74 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NemoriLLM=GPT-4o-mini2026.01 | 44.72 | — | 46.33 | 38.52 | — | 32.36 | 23.36 | — | 55.99 | 48.25 | — | 29.19 | 22.31 | — | 36.75 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| NemoriTokens=4767, Latency (s)=2.5122026.01 | 44.72 | — | — | — | — | — | — | — | — | — | — | — | — | — | 36.75 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LangMem2026.02 | 44.32 | — | 51.13 | 44.22 | — | 41.11 | 32.09 | — | 53.67 | 46.22 | — | 33.38 | 27.26 | — | 37.95 | 37.71 | 47.37 | 24.55 | 48.37 | 39.5 | — | — | — | — | — | — | — | — | — | — | — | |
| FullContextBackbone=Qwen2-7B-Instruct2026.01 | 44.3 | 0.803 | 50.2 | 43 | 0.632 | 35.1 | 25.9 | 0.527 | 42.1 | 35.1 | 0.523 | 25.9 | 17.6 | 0.696 | 37.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ZepLLM=GPT-4o-mini2026.01 | 43.56 | — | 49.56 | 38.92 | — | 35.74 | 23.3 | — | 42 | 34.53 | — | 19.37 | 14.82 | — | 33.64 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ZepTokens=3911, Latency (s)=1.2922026.01 | 43.56 | — | — | — | — | — | — | — | — | — | — | — | — | — | 33.64 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemVerseBackbone=GPT-4o-mini2026.04 | 43.44 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimpleMemBackbone=GPT-4o2026.04 | 43.2 | — | 19.5 | — | — | 31.8 | — | — | 23.5 | — | — | 30.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | 80.2 | — | — | — | — | — | |
| MemoryOSLLM=GPT-4o-mini2026.01 | 42.84 | — | 48.62 | 42.99 | — | 35.27 | 25.22 | — | 41.15 | 30.76 | — | 20.02 | 16.52 | — | 35.53 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemoryOSTokens=3874, Latency (s)=8.6572026.01 | 42.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | 35.53 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemoryOSBackbone=GPT-4o-mini2026.04 | 42.84 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem0Backbone=GPT-4o2026.01 | 42.4 | 0.697 | 47.4 | 41 | 0.666 | 39.1 | 29.5 | 0.555 | 38.2 | 32.3 | 0.468 | 23.2 | 17.3 | 0.647 | 35.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem0Backbone=GPT-4o-mini2026.04 | 42.13 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimpleMemBackbone=GPT-5.12026.04 | 41.8 | — | 17.8 | — | — | 30.5 | — | — | 27.2 | — | — | 30.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | 80.7 | — | — | — | — | — | |
| MemoryOS2026.02 | 41.28 | — | 49.31 | 42.61 | — | 43.03 | 34.09 | — | 44.43 | 32.03 | — | 28.36 | 23.24 | — | 32.99 | 39.75 | 42.64 | 27.83 | 49.84 | 40.02 | — | — | — | — | — | — | — | — | — | — | — | |
| Mem0Backbone=GPT-4o-mini2026.01 | 41.2 | 0.676 | 44.1 | 34 | 0.598 | 34 | 25 | 0.5 | 44 | 37.3 | 0.403 | 26.9 | 19.2 | 0.608 | 33.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemGPTBackbone=GPT-4o2026.04 | 40.4 | — | 18.8 | — | — | 30.5 | — | — | 24.6 | — | — | 30.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | 84.3 | — | — | — | — | — | |
| SimpleMemBackbone=GPT-4o-mini2026.04 | 40.4 | — | 12.8 | — | — | 30 | — | — | 17.8 | — | — | 31.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | 89.1 | — | — | — | — | — | |
| Mem0Backbone=GPT-4o2026.04 | 39.7 | — | 15.6 | — | — | 30.9 | — | — | 21.7 | — | — | 29.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | 85.7 | — | — | — | — | — | |
| A-MemLLM=GPT-4o-mini2026.01 | 39.65 | — | 44.65 | 37.06 | — | 27.02 | 20.09 | — | 45.85 | 36.67 | — | 12.14 | 12 | — | 32.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| A-MemTokens=2712, Latency (s)=1.412026.01 | 39.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | 32.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| A-MEMBackbone=GPT-4o2026.04 | 39.4 | — | 17.4 | — | — | 29.5 | — | — | 20 | — | — | 26.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | 89.8 | — | — | — | — | — | |
| Mem0Backbone=GPT-5.12026.04 | 39 | — | 16 | — | — | 29.2 | — | — | 26.1 | — | — | 29.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | 81.9 | — | — | — | — | — | |
| Claude-MemBackbone=GPT-5.12026.04 | 38.8 | — | 17.1 | — | — | 28.9 | — | — | 26.4 | — | — | 29.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | 81.4 | — | — | — | — | — | |
| SimpleMemBackbone=GPT-5-nano2026.04 | 38.8 | — | 20 | — | — | 27.8 | — | — | 24.5 | — | — | 28.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | 82.4 | — | — | — | — | — | |
| A-MEMBackbone=GPT-5.12026.04 | 38.5 | — | 16.4 | — | — | 28.7 | — | — | 24.6 | — | — | 28.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | 82.6 | — | — | — | — | — | |
| MemGPTBackbone=GPT-5.12026.04 | 38.5 | — | 16.5 | — | — | 28.8 | — | — | 24.9 | — | — | 29.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | 80.6 | — | — | — | — | — | |
| Claude-MemBackbone=GPT-4o2026.04 | 38.3 | — | 15.3 | — | — | 29.4 | — | — | 16.7 | — | — | 24.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | 91.5 | — | — | — | — | — | |
| MemVerseBackbone=GPT-5.12026.04 | 38.3 | — | 17.3 | — | — | 28.7 | — | — | 27.7 | — | — | 29.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | 78 | — | — | — | — | — | |
| ZepBackbone=GPT-4o-mini2026.01 | 37.2 | 0.627 | 39.4 | 33.4 | 0.501 | 27.3 | 19.2 | 0.584 | 44.4 | 37.8 | 0.393 | 22.7 | 15.6 | 0.58 | 30.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LangMemLLM=GPT-4o-mini2026.01 | 36.87 | — | 40.91 | 33.63 | — | 35.51 | 26.86 | — | 30.75 | 25.84 | — | 26.04 | 22.32 | — | 30.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LangMemTokens=127, Latency (s)=18.532026.01 | 36.87 | — | — | — | — | — | — | — | — | — | — | — | — | — | 30.06 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemVerseBackbone=GPT-5-nano2026.04 | 36.6 | — | 20.3 | — | — | 20.8 | — | — | 16.8 | — | — | 25.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | 74.1 | — | — | — | — | — | |
| LangMemBackbone=Qwen2-72B-Instruct2026.01 | 36.5 | 0.633 | 39 | 33.7 | 0.529 | 35.3 | 25.4 | 0.253 | 32.5 | 27 | 0.545 | 34.1 | 24.9 | 0.525 | 30.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemVerseBackbone=GPT-4o2026.04 | 36.5 | — | 15.7 | — | — | 26 | — | — | 19.6 | — | — | 19.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | 94.4 | — | — | — | — | — | |
| Mem0Backbone=GPT-4o-mini2026.04 | 36.4 | — | 11.2 | — | — | 28.5 | — | — | 17.9 | — | — | 29.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | 76.1 | — | — | — | — | — | |
| MemGPTBackbone=GPT-4o-mini2026.04 | 36.4 | — | 11.3 | — | — | 28.3 | — | — | 18.2 | — | — | 28.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | 77.6 | — | — | — | — | — | |
| ZepBackbone=GPT-4o2026.01 | 36 | 0.652 | 44.4 | 39 | 0.523 | 29.7 | 19.9 | 0.587 | 23.3 | 19.5 | 0.427 | 23.6 | 18.8 | 0.601 | 30.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LangMemBackbone=GPT-4o-mini2026.01 | 35.9 | 0.616 | 38.9 | 33.2 | 0.526 | 33.6 | 24 | 0.25 | 32.1 | 26.3 | 0.477 | 29.5 | 23.6 | 0.515 | 29.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| A-MEMBackbone=GPT-4o-mini2026.04 | 35.7 | — | 9.1 | — | — | 27.8 | — | — | 16.3 | — | — | 26 | — | — | — | — | — | — | — | — | — | — | — | — | — | 82.3 | — | — | — | — | — | |
| LangMemBackbone=Qwen2-7B-Instruct2026.01 | 35.6 | 0.617 | 38 | 32.8 | 0.515 | 34.4 | 24.7 | 0.246 | 31.7 | 26.3 | 0.531 | 33.2 | 24.3 | 0.512 | 30.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemGPTBackbone=GPT-5-nano2026.04 | 35.5 | — | 15.1 | — | — | 26.7 | — | — | 22.6 | — | — | 27.1 | — | — | — | — | — | — | — | — | — | — | — | — | — | 74.4 | — | — | — | — | — | |
| Mem0Backbone=GPT-5-nano2026.04 | 35.2 | — | 14.3 | — | — | 26.4 | — | — | 23.7 | — | — | 27 | — | — | — | — | — | — | — | — | — | — | — | — | — | 73.7 | — | — | — | — | — | |
| A-MEMBackbone=GPT-5-nano2026.04 | 34.8 | — | 14.9 | — | — | 26 | — | — | 22.3 | — | — | 25.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | 74.5 | — | — | — | — | — | |
| SimpleMemBackbone=GPT-4.1-nano2026.04 | 34.2 | — | 14.5 | — | — | 29.8 | — | — | 21 | — | — | 28.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | 64.8 | — | — | — | — | — | |
| Claude-MemBackbone=GPT-4o-mini2026.04 | 33.8 | — | 10.2 | — | — | 24.5 | — | — | 12.2 | — | — | 21.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | 84.5 | — | — | — | — | — | |
| MemGPTLLM=GPT-4o-mini2026.01 | 33.17 | — | 41.01 | 34.34 | — | 26.65 | 17.72 | — | 25.52 | 19.44 | — | 9.15 | 7.44 | — | 26.51 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| A-Mem2026.02 | 31.69 | — | 40.94 | 35.54 | — | 32.41 | 24.28 | — | 37.23 | 33.02 | — | 16.16 | 14.64 | — | 26.87 | 31.26 | 39.61 | 15.19 | 42.32 | 32.1 | — | — | — | — | — | — | — | — | — | — | — | |
| MemGPTBackbone=GPT-4.1-nano2026.04 | 31.6 | — | 13 | — | — | 28.7 | — | — | 23.4 | — | — | 27.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | 55.6 | — | — | — | — | — | |
| MemoryBankBackbone=GPT-4o-mini2026.04 | 31.42 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Mem0Backbone=GPT-4.1-nano2026.04 | 31 | — | 13.4 | — | — | 29 | — | — | 19.4 | — | — | 27.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | 53.7 | — | — | — | — | — | |
| ES-MemLLM=Qwen2.5-3B2026.01 | 30.86 | — | 35.47 | 30.54 | — | 21.68 | 15.18 | — | 31.48 | 26.22 | — | 15.37 | 11.56 | — | 25.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ES-MemLLM=Llama3.2-3B2026.01 | 30.46 | — | 35.18 | 30.59 | — | 26.78 | 20.53 | — | 25.83 | 20.69 | — | 15.42 | 11.15 | — | 25.48 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| A-MemBackbone=GPT-4o2026.01 | 29.3 | 0.53 | 27.6 | 20.6 | 0.254 | 12.4 | 12.3 | 0.559 | 47 | 37.5 | 0.553 | 34 | 28.2 | 0.487 | 23 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MemVerseBackbone=GPT-4o-mini2026.04 | 29 | — | 7.4 | — | — | 14.7 | — | — | 10.6 | — | — | 9.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | 74.7 | — | — | — | — | — | |
| A-MemBackbone=GPT-4o-mini2026.01 | 28.6 | 0.518 | 27 | 20.1 | 0.248 | 12.1 | 12 | 0.546 | 45.9 | 36.7 | 0.541 | 33.3 | 27.6 | 0.476 | 22.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| A-MemBackbone=Qwen2-72B-Instruct2026.01 | 28.6 | 0.518 | 27 | 20.1 | 0.248 | 12.1 | 12 | 0.546 | 45.9 | 36.7 | 0.541 | 33.3 | 27.6 | 0.476 | 22.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — |