STEM Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
70.02Accuracy
23
Mar 12, 2026
69.96Accuracy
23
Mar 12, 2026
32Pass@1 Accuracy
16
Jun 25, 2026
70.2Accuracy
16
Mar 4, 2026
55.4Avg@2
16
Feb 26, 2026
73.7Accuracy (STEM)
15
May 21, 2026
2.096Speedup Ratio
15
Mar 26, 2026
91.9Pass@1
15
Mar 24, 2026
77.06Pass@1 Accuracy
14
Jun 25, 2026
55Accuracy (avg@4)
12
Mar 30, 2026
82.1AUROC
10
Mar 20, 2026
57.3Accuracy
10
Feb 26, 2026
47.6Accuracy
9
Jul 2, 2026
97.26Pass@1
8
May 29, 2026
87.8Pass@1
8
May 29, 2026
90Pass@1
8
May 29, 2026
96.81Pass@1
8
May 29, 2026
97.77Pass@1 Accuracy
8
May 29, 2026
90.8Pass@1 Accuracy
8
May 29, 2026
36.8Accuracy
8
Apr 7, 2026
56.57Avg@32
8
Mar 24, 2026
32Accuracy
7
Jun 4, 2026
41.1Accuracy
7
May 21, 2026
25.9Accuracy
7
May 21, 2026
0.893Accuracy
7
May 21, 2026