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

Mathematical Reasoning on Math Benchmarks Average

76.1Accuracy (ACC)

GPT-5-Thinking*

16.92432.28747.6563.013Oct 2, 2025Nov 10, 2025Dec 19, 2025Jan 28, 2026Mar 8, 2026Apr 16, 2026May 26, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
76.1-
2026.04
76-
2025.10
69.325,718.54
2025.10
68.595,806.23
2025.10
67.837,317.59
2025.10
67.785,470.29
2025.10
67.67,529.24
2026.04
60.954,660
2025.10
60.815,821.54
2026.04
60.155,972
2026.04
59.728,134
2026.04
59.365,515
2026.04
58.86,983
2026.04
58.257,787
2026.04
55.996,964
2026.04
54.747,810
2026.04
54.527,958
2026.04
54.117,260
2026.04
51.6-
2026.05
50.2-
2026.04
49.914,762
2026.05
49.5-
2026.05
49.2-
2026.04
49.1-
2025.10
48.982,172.07
2025.10
48.82,293.45
2026.04
48.785,574
2025.10
48.591,435.86
2026.05
48.3-
2026.05
48.1-
2025.10
47.81,452.11
2026.05
47.3-
2026.04
47.2-
2026.05
47.1-
2026.04
46.687,442
2026.05
46.3-
2026.04
46.1-
2026.04
46.092,806
2026.05
45.8-
2026.05
43.1-
2026.04
43-
2026.04
41.7-
2026.04
41.568,110
2026.04
40.78,273
2026.04
40.1-
2026.05
22.7-
2026.05
19.2-