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

Mathematical Problem Solving on AIME 2025

100Score

GPT-5.2

-3.79223.15450.177.046Dec 15, 2025Jan 9, 2026Feb 4, 2026Mar 2, 2026Mar 28, 2026Apr 23, 2026May 19, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.05
100
2026.05
100
2026.05
98.33
2026.05
97.5
2026.05
96.67
2026.05
96.67
96.04
2026.05
95.83
95.83
2026.05
95.83
95.73
2026.05
95.42
95
2026.05
94.06
2026.05
94.06
93.44
2026.05
93.44
93.33
93.02
92.92
2026.05
91.88
2026.01
91.7
2026.05
91.04
2026.05
90.94
2026.05
90.94
2026.05
90.31
2026.05
89.17
2026.05
88.96
88.96
88.65
2026.05
87.92
2026.05
86.77
86.25
2026.05
86.04
2026.05
85.83
2026.05
85.42
85.1
84.79
2026.01
84.3
2026.05
83.85
2026.05
82.81
2026.01
82.7
2026.05
82.4
81.04
2026.05
80
2026.05
79.06
2026.05
78.96
2026.05
76.15
2026.01
75
2026.05
70.21
2026.05
69.48
2026.05
69.17
2026.05
68.96
2026.05
68.65
2026.05
65.94
2026.05
65.73
2026.05
63.75
61.15
61.04
2026.05
49.06
2026.05
46.88
45.1
2025.12
43.3
35.31
33.44
2026.05
32.6
2025.12
32.5
22.4
2025.12
21.7
2025.12
20.4
2026.05
15.21
2025.12
7.2
2025.12
6.3
2025.12
6.3
2025.12
0.4
2025.12
0.2