Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
256.28Inference Speed (TPS)
20
Feb 26, 2026
66.91AUROC
20
Feb 26, 2026
0.621AUROC
20
Feb 26, 2026
0.651AUROC
20
Feb 26, 2026
86.4Pass@1
20
Feb 26, 2026
85.4Accuracy
20
Feb 26, 2026
91.3Accuracy
20
Feb 27, 2026
54Accuracy
20
Feb 27, 2026
92.7Accuracy
20
Feb 26, 2026
26.5Accuracy
20
Feb 26, 2026
33.6Accuracy
20
Apr 2, 2026
93.3Accuracy
19
Jul 1, 2026
94.34Pass@1 Accuracy
19
Jun 9, 2026
84.4Accuracy
19
Jun 30, 2026
91.74Pass@1 Accuracy
19
Jun 9, 2026
53.84Pass@32 Accuracy
19
Jul 7, 2026
54.06AIME24 Avg@32 Score
19
May 29, 2026
90.62Avg@16
19
May 29, 2026
1.93Speedup (vs. AR)
19
May 27, 2026
91.6Arithmetic Accuracy
19
Jun 30, 2026
74.38Accuracy
19
May 26, 2026
99.15Accuracy
19
May 18, 2026
19.5AIME 2024 Score
19
May 13, 2026
72.5Success Rate (SR)
19
Apr 23, 2026
43.9Avg@4
19
Apr 14, 2026