Long-term Agent Memory Evaluation on LongMemEval
95.7SS-UTiMem
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| TiMemEvaluation Protocol=TiMem (GPT-4o-mini, LLJ accuracy)2026.05 | 95.7 | 82.1 | 89.7 | 63.3 | 86.2 | 68.4 | 70.8 | 76.9 | |
| Mem0Evaluation Protocol=TiMem (GPT-4o-mini, LLJ accuracy)2026.05 | 94.3 | 51.8 | 75.4 | 50 | 78.7 | 49.2 | 66.2 | 65 | |
| MemOSEvaluation Protocol=TiMem (GPT-4o-mini, LLJ accuracy)2026.05 | 93.7 | 67.9 | 82.2 | 50.7 | 76.7 | 65.1 | 58.8 | 68.7 | |
| ZepEvaluation Protocol=MemOS evaluation suite (short-answer prompt)2026.05 | 92.9 | 75 | 84.9 | 53.3 | 74.4 | 54.1 | 47.4 | 63.2 | |
| memobaseEvaluation Protocol=MemOS evaluation suite (short-answer prompt)2026.05 | 92.8 | 23.2 | 61.9 | 80 | 89.7 | 75.9 | 66.9 | 72.4 | |
| SupermemoryEvaluation Protocol=MemOS evaluation suite (short-answer prompt)2026.05 | 85.7 | 58.9 | 73.8 | 90 | 55.1 | 44.4 | 52.6 | 58.4 | |
| A-MEMEvaluation Protocol=TiMem (GPT-4o-mini, LLJ accuracy)2026.05 | 82.9 | 87.5 | 84.9 | 39.3 | 72.8 | 36.1 | 40.3 | 55.4 | |
| Mem0Evaluation Protocol=MemOS evaluation suite (short-answer prompt)2026.05 | 82.9 | 26.8 | 57.9 | 90 | 66.7 | 72.2 | 63.1 | 66.4 | |
| MemoryOSEvaluation Protocol=TiMem (GPT-4o-mini, LLJ accuracy)2026.05 | 81.1 | 78.2 | 79.8 | 51.3 | 56.1 | 53.4 | 44.8 | 58.1 | |
| SAGEZero-shot=false, Training Status=trained, Interaction Rounds=+1 round2026.05 | 79.4 | 80.5 | 79.6 | 15.8 | 26.2 | 22.8 | 10.7 | 34.1 | |
| SAGEZero-shot=false, Training Status=trained2026.05 | 73.3 | 80 | 76 | 23.1 | 27.8 | 22.8 | 12.5 | 34.3 | |
| MIRIXEvaluation Protocol=MemOS evaluation suite (short-answer prompt)2026.05 | 72.8 | 63.6 | 68.8 | 53.3 | 52.6 | 25.6 | 30.1 | 43.5 | |
| MemUEvaluation Protocol=MemOS evaluation suite (short-answer prompt)2026.05 | 67.1 | 19.6 | 46 | 76.7 | 41 | 17.3 | 42.1 | 38.4 | |
| SAGEZero-shot=true, Training Status=untrained2026.05 | 60.3 | 81.4 | 68 | 16.7 | 23.5 | 13.8 | 9.4 | 28.4 | |
| MemoryBankEvaluation Protocol=TiMem (GPT-4o-mini, LLJ accuracy)2026.05 | 50 | 9.8 | 32.1 | 0 | 21.8 | 0 | 0 | 11.5 |