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

Mathematical Problem Solving on AIME 2024 (Pass@k %)

63Pass@k

GCPO

14.43227.04139.6552.259May 12, 2026
Updated 21d ago

Evaluation Results

MethodLinks
2026.05
63
2026.05
62.2
2026.05
59.1
2026.05
57.3
2026.05
56.3
2026.05
54.2
2026.05
52.8
2026.05
49.6
2026.05
49.4
2026.05
47.2
2026.05
43.7
2026.05
40.7
2026.05
40.6
2026.05
39.1
2026.05
37.4
2026.05
34.5
2026.05
33.5
2026.05
30.2
2026.05
30
2026.05
28.1
2026.05
25.2
2026.05
24.6
2026.05
20.9
2026.05
16.3