Conversational Memory on LongMemEval 1.0 (test)
92.6Overall AccuracyChronos Low
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Chronos Low2026.03 | 92.6 | 96.15 | 91.73 | 100 | 80 | 94.29 | 90.23 | |
| HonchoEvaluation Model=Claude Haiku 4.52026.03 | 90.4 | 94.87 | 84.96 | 96.43 | 90 | 94.29 | 88.72 | |
| EmergenceMem Internal2026.03 | 86 | 83.33 | 81.2 | 100 | 60 | 98.57 | 85.71 | |
| Mastra2026.03 | 84.8 | 85.9 | 79.7 | 82.14 | 73.33 | 98.57 | 85.71 | |
| HindsightActor Model=OSS-20B, Judge Model=OSS-120B2026.03 | 83.6 | 84.6 | 79.7 | 94.6 | 66.7 | 95.7 | 79.7 | |
| Supermemory2026.03 | 81.6 | 88.5 | 71.4 | 96.4 | 70 | 97.1 | 76.7 | |
| Zep2026.03 | 71.2 | 83.3 | 57.9 | 80.4 | 56.7 | 92.9 | 62.4 | |
| Full-context2026.03 | 60.2 | 78.2 | 44.3 | 94.6 | 20 | 81.4 | 45.1 |