Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
83.82Pass@8 Accuracy
5
May 12, 2026
63.33Pass@8 Score
5
May 12, 2026
83.33Pass@8
5
May 12, 2026
23.4Accuracy
5
May 12, 2026
65.5Average Score
5
May 11, 2026
36.7Peak Accuracy
5
May 11, 2026
81.2CR (%)
5
May 8, 2026
93.3Accuracy
5
May 7, 2026
13.8Accuracy
5
May 5, 2026
58.32Pass@1
5
May 1, 2026
60.78Pass@1
5
May 1, 2026
15.83Pass@1
5
May 1, 2026
23.23Pass@1 Rate
5
May 1, 2026
26.56Pass@1 Accuracy
5
May 1, 2026
80.4Pass@1
5
May 1, 2026
85.34RPR
5
Apr 30, 2026
81.5Accuracy
5
Apr 30, 2026
78.42Accuracy (GSM8K en)
5
Apr 30, 2026
53.3Token Savings
5
Apr 29, 2026
100Accuracy under Attack (GSM8K)
5
Apr 28, 2026
62.1Accuracy
5
Apr 28, 2026
57Accuracy
5
Apr 28, 2026
79.8Accuracy
5
Apr 28, 2026
37.3Accuracy
5
Apr 28, 2026
49.2Accuracy
5
Apr 28, 2026