Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Multi-step mathematical reasoning on We-Math (test)

72.8S1 Score

GPT-4o

30.78441.69252.663.508May 15, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.05
72.858.143.6
2025.05
65.549.238.2
2025.05
65.458.652.1
2025.05
65.249.449.1
2025.05
59.443.635.2
2025.05
59.143.626.7
2025.05
57.237.934.6
2025.05
56.151.433.9
2025.05
53.339.436.4
2025.05
5242.238.8
2025.05
5139.246.1
2025.05
4733.133.3
2025.05
45.638.133.9
2025.05
44.435.327.9
2025.05
40.830.320.6
2025.05
38.734.234.6
2025.05
35.43032.7
2025.05
34.326.120
2025.05
32.626.725.5
2025.05
32.430.632.7