Agent Memory on AMA-Bench real-world
62.38Recall (Accuracy)
AMA-Agent
Updated 3mo ago
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
2026.02 | 62.38 | 32.8 | 61.45 | 31.03 | 53.05 | 26.25 | 47.19 | 28.25 | 57.22 | 29.92 | |
2026.02 | 48.43 | 15.9 | 49.74 | 15.49 | 35.2 | 13.53 | 30.11 | 16.1 | 42.27 | 15.22 | |
2026.02 | 47.08 | 17.89 | 54.97 | 18.11 | 42.57 | 17.13 | 36.59 | 20.73 | 46.06 | 18.22 | |
2026.02 | 45.79 | 23.56 | 50.8 | 19.66 | 44.03 | 18.92 | 35.38 | 17.85 | 44.8 | 20.48 | |
2026.02 | 33.01 | 14.65 | 42.64 | 15.49 | 34.5 | 13.25 | 24.98 | 16.23 | 34.36 | 14.75 | |
2026.02 | 32.89 | 13.18 | 44.04 | 14.75 | 28.09 | 12.59 | 25.26 | 14.31 | 33.04 | 13.59 | |
2026.02 | 32.31 | 31.28 | 41 | 28.61 | 30.06 | 26.78 | 33.32 | 30.11 | 33.97 | 29.28 | |
2026.02 | 30.84 | 27.07 | 36.53 | 27.31 | 30.88 | 24.8 | 28.73 | 29.53 | 31.86 | 26.95 | |
2026.02 | 30.77 | 27.69 | 39.05 | 26.34 | 31.4 | 25.51 | 28.79 | 25.88 | 32.58 | 26.5 | |
2026.02 | 28.76 | 23.25 | 41.72 | 19.93 | 30.64 | 20 | 21.71 | 21.35 | 31.17 | 21.3 | |
2026.02 | 25.5 | 14.89 | 33.8 | 16.06 | 28.49 | 14.32 | 22.02 | 16.55 | 27.68 | 15.3 | |
2026.02 | 20.12 | 20.39 | 18.84 | 16.12 | 17.64 | 15.94 | 13.73 | 16.89 | 18.11 | 17.64 | |
2026.02 | 20.11 | 24.13 | 26.45 | 24.43 | 21.01 | 22.25 | 15.16 | 22.41 | 21.04 | 23.43 | |
2026.02 | 11.8 | 18.57 | 14.27 | 17.32 | 12.05 | 16.59 | 10.8 | 20.42 | 12.29 | 18.07 |