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

Reasoning on GSM8K (test)

89.6EM Accuracy

MCTS

47.11658.145569.17580.2045Mar 19, 2024Jun 30, 2024Oct 11, 2024Jan 23, 2025May 6, 2025Aug 17, 2025Nov 29, 2025
Updated 3d ago

Evaluation Results

MethodLinks
2025.02
89.679.68
2025.02
89.419.95
2025.02
88.234.89
2025.02
86.736.49
2025.11
81.2-
2025.11
79.5-
2024.03
79.08-
2024.03
79.08-
2024.03
78.92-
2024.03
78.85-
2024.03
77.79-
2025.11
77.6-
2024.03
77.48-
2024.03
77.41-
2025.11
77-
2025.11
76.2-
2025.11
75.8-
2025.02
75.218.32
2025.02
75.173.28
2025.02
71.541.27
2025.02
70.133.37
2025.11
69.8-
2025.02
69.569.74
2025.02
68.334.51
2025.02
68.217.82
2025.11
68.1-
2025.02
67.827.74
2025.02
67.633.48
2025.11
67.3-
2025.02
6457.19
2025.02
58.428.27
2025.02
57.127.27
2024.03
53.98-
2024.03
52.99-
2024.03
48.75-