Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1Total Inference Runtime (s)
36
Apr 6, 2026
0Total Inference Runtime (mm:ss)
36
Apr 6, 2026
37.86EM Score
36
May 11, 2026
73.33Accuracy (%)
36
May 13, 2026
78.4Accuracy
36
Jun 2, 2026
40.24Accuracy
36
Mar 24, 2026
97.8Accuracy
36
Mar 17, 2026
94.77Accuracy (avg@32)
36
Jun 17, 2026
54.04Accuracy
36
Apr 20, 2026
96Accuracy
36
Feb 26, 2026
90.22Accuracy
36
Feb 26, 2026
92.48Accuracy
36
Feb 26, 2026
90.66Accuracy
36
Feb 26, 2026
65.2Score
36
May 22, 2026
89.22Avg@16
36
Feb 27, 2026
73.4Accuracy
36
May 5, 2026
60.67Pass Rate
36
Feb 26, 2026
68.5Accuracy
36
Feb 26, 2026
57.9EM
36
Mar 4, 2026
76.6Accuracy
36
Apr 23, 2026
85.1Accuracy
36
Feb 26, 2026
79.9Accuracy
35
Jun 30, 2026
81.7Accuracy
35
Jun 30, 2026
49.96Flexible Extract Score
35
May 22, 2026
35.2Accuracy
35
Jun 16, 2026