Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
79.4Accuracy
38
Jun 16, 2026
35.69AIME Accuracy
38
Apr 20, 2026
87.97Accuracy
38
May 29, 2026
56.5Accuracy
38
Jun 18, 2026
33.54Pass@1
38
May 22, 2026
47Pass@1
38
Mar 4, 2026
76.92Accuracy
38
Feb 26, 2026
97.8Pass@1
38
Apr 14, 2026
84.5EM
38
Feb 26, 2026
78.4GSM8K Score
38
Jun 2, 2026
15,508Mean Peak GPU Memory (MB)
37
Jun 26, 2026
74.2Accuracy
37
Jun 26, 2026
93.5Accuracy
37
Apr 16, 2026
75Pass@1
37
Mar 4, 2026
69.2Accuracy
37
Apr 10, 2026
90.57Accuracy
37
Feb 26, 2026
3.12Avg Acceptance Length (τ)
36
Jul 7, 2026
84Mean Accuracy
36
Jun 1, 2026
98.2Mean Accuracy (MATH500)
36
Jun 1, 2026
86.7Pass@16
36
Jun 9, 2026
33.2Avg@16 Accuracy
36
Jun 9, 2026
45.5Accuracy
36
May 14, 2026
97Accuracy
36
Apr 10, 2026
99.4Overall Result Accuracy
36
Jun 23, 2026
0Total Inference Time (mm:ss)
36
Apr 6, 2026