Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
12.73Accuracy
7
Mar 9, 2026
88.71Accuracy
7
Feb 26, 2026
83.4BBH
7
Feb 26, 2026
77.2Ko-Winogrande
7
Feb 26, 2026
98.3AIME 2025 Score
6
Jul 7, 2026
92.4AIME Score
6
Jul 7, 2026
28.41Precision@1
6
Jun 5, 2026
56.59LLM-as-a-judge Score
6
Jun 4, 2026
100Success Rate
6
Jun 1, 2026
100Success Rate
6
Jun 1, 2026
74.8MMLU-Pro Accuracy
6
May 28, 2026
61.8Average Score
6
May 21, 2026
93.7Graph SR
6
May 20, 2026
80.5Accuracy
6
May 14, 2026
47.2Accuracy
6
May 4, 2026
100Accuracy (Level 1)
6
Apr 23, 2026
97.9Accuracy
6
Apr 10, 2026
83.9Accuracy
6
Apr 6, 2026
70.6Normalized Log Accuracy
6
Mar 18, 2026
42.1Pass@1
6
Mar 4, 2026
88.02Correctness
6
Feb 26, 2026
94.18Correctness
6
Feb 26, 2026
84.69Correctness
6
Feb 26, 2026
47.34Correctness
6
Feb 26, 2026
28.9Delta Accuracy
6
Feb 26, 2026