Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
96.3Avg@32
24
May 12, 2026
46.9Accuracy (GSM8K)
24
Apr 10, 2026
57.1Pass@1 Accuracy
24
Jun 23, 2026
48.4Pass@1
24
Apr 6, 2026
60.43Pass@4
24
Apr 3, 2026
80.01Solved Rate
24
May 29, 2026
30.6Accuracy (VisuLogic Math)
24
Apr 21, 2026
29.6Mean Accuracy
24
Mar 6, 2026
59.5Mean Accuracy
24
Mar 6, 2026
48.67Exact Match (EM)
24
Mar 4, 2026
78.4Accuracy
24
May 27, 2026
0.44Accuracy
24
Feb 26, 2026
81Accuracy
24
Feb 26, 2026
68.66Accuracy
24
Feb 26, 2026
40Accuracy
24
Feb 26, 2026
54Accuracy
24
Feb 26, 2026
93Accuracy
24
Feb 26, 2026
92.6Accuracy
24
Feb 26, 2026
80Accuracy
24
Apr 14, 2026
83.02Accuracy
24
Feb 26, 2026
34.38Avg@32
24
Feb 26, 2026
54.06Avg@32
24
Feb 26, 2026
88.05Avg@16 Score
24
Feb 26, 2026
14.8Average@16
24
Jun 9, 2026
14.2Inference Latency (s)
24
Feb 26, 2026