Open-ended scientific discovery on DiscoveryBench (evaluation set)
4.91ESOpenEvolve
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| OpenEvolve2026.06 | 4.91 | 1.92 | 68 | 12 | 17.6 | |
| Raw agent2026.06 | 4.26 | 2.61 | 54 | 7 | 13 | |
| SAGA2026.06 | 3.9 | 2.59 | 45 | 4 | 8.9 | |
| StatefulDiscovery2026.06 | 3.41 | 3.39 | 39 | 9 | 23.1 | |
| AutoDiscovery2026.06 | 3.39 | 2.9 | 59 | 7 | 11.9 |