Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Mathematical and Physics Reasoning on OLYMPIADBENCH

65.15Accuracy

SERO

33.669241.842150.01558.1879May 27, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2026.05
65.15
2026.05
60.74
2026.05
55.97
2026.05
52.97
2026.05
51.02
2026.05
48.6
2026.05
47.89
2026.05
47.54
2026.05
42.9
2026.05
40.93
2026.05
40.92
2026.05
40.27
2026.05
40.07
2026.05
39.61
2026.05
38.97
2026.05
38.54
2026.05
38.46
2026.05
37.71
2026.05
36.6
2026.05
35.57
2026.05
34.88