Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-ended scientific discovery on DiscoveryBench (evaluation set)
Loading...
4.91
ES
OpenEvolve
3.3292
3.7396
4.15
4.5604
Jun 10, 2026
ES
DV
Claims Count
HQ Count
HQ Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
ES
DV
Claims Count
HQ Count
HQ Rate
OpenEvolve
2026.06
4.91
1.92
68
12
17.6
Raw agent
2026.06
4.26
2.61
54
7
13
SAGA
2026.06
3.9
2.59
45
4
8.9
StatefulDiscovery
2026.06
3.41
3.39
39
9
23.1
AutoDiscovery
2026.06
3.39
2.9
59
7
11.9
Feedback
Search any
task
Search any
task