Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
40AIME 24 Score
7
Jun 4, 2026
61.6Accuracy
7
Jun 4, 2026
79.3Accuracy
7
Jun 4, 2026
62.6Accuracy
7
Jun 4, 2026
58.2Accuracy
7
Jun 4, 2026
77.5Accuracy
7
Jun 4, 2026
99.1Success Rate
7
Jun 1, 2026
91.1HellaSwag Score
7
May 13, 2026
82.8FA
7
Apr 24, 2026
95.9Accuracy (FA)
7
Apr 24, 2026
58.3FA Score (WinoGrande)
7
Apr 24, 2026
77.2Factual Accuracy (FA)
7
Apr 24, 2026
79.05Pass@1 (MMLU STEM Reasoning)
7
Apr 21, 2026
87.2Pass@1
7
Apr 21, 2026
21.23BLEU
7
Mar 26, 2026
21.83BLEU
7
Mar 26, 2026
78EM
7
Mar 20, 2026
75.8Accuracy
7
Mar 31, 2026
23.89Accuracy (GPQA)
7
Mar 17, 2026
92.84Accuracy (%)
7
May 11, 2026
61.5Accuracy
7
Mar 11, 2026
75Accuracy
7
Mar 11, 2026
85.5Accuracy
7
Mar 11, 2026
86Accuracy
7
Mar 11, 2026
89.5Accuracy
7
Mar 11, 2026