Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
99.4Score
22
Jul 7, 2026
81.5Accuracy
22
Feb 26, 2026
58.71Accuracy
22
May 12, 2026
2.44Normalized Speedup
21
Jun 5, 2026
2.68Normalized Speedup
21
Jun 5, 2026
3.15Normalized Speedup
21
Jun 5, 2026
80.85ARC-E Accuracy
21
Jun 5, 2026
93.9ARC-c Accuracy (avg@4)
21
Jun 4, 2026
85.77Average Score
21
Apr 27, 2026
74.4Score (%)
21
Apr 20, 2026
32.8Accuracy
21
May 12, 2026
79.1Accuracy
21
Apr 8, 2026
99.8Pass@1 Accuracy
21
May 21, 2026
90.71Accuracy
21
Feb 26, 2026
60.83Accuracy
21
Feb 26, 2026
90.63Accuracy (AMC 2023)
21
Feb 26, 2026
92.2p_T Score
21
Feb 26, 2026
97.1p_T Score
21
Feb 26, 2026
82.15T3.1 (CoT)
21
Feb 26, 2026
90.58Pass@1
21
Feb 26, 2026
61.3Score
21
Feb 26, 2026
79.8Score
21
Feb 26, 2026
67.07Score
21
Feb 26, 2026
81.1Score
21
Feb 26, 2026
76.25Reasoning Score
21
Feb 26, 2026