Scientific Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
83.41Score
9
Feb 26, 2026
10.54Score
9
Feb 26, 2026
48.4Accuracy
9
Feb 26, 2026
67.42Accuracy
8
Jun 12, 2026
68.2Pass@1 Accuracy
8
Jun 8, 2026
65.9Expert Count
8
Jun 1, 2026
33.3Score
8
Jul 2, 2026
34.3Accuracy
8
May 20, 2026
8.93BLEU
8
May 7, 2026
19.9Accuracy
8
Apr 14, 2026
65.7Accuracy
8
Apr 14, 2026
63.7Accuracy
8
Apr 14, 2026
56.9Accuracy
8
Apr 14, 2026
28.36Chemistry (Pro)
8
May 29, 2026
51.01Overall Score
8
Mar 20, 2026
79Avg@8
8
Mar 18, 2026
14.7ROUGE-L
8
Mar 4, 2026
76.82Accuracy
8
Feb 26, 2026
28.57Pass@1 Accuracy
8
Feb 26, 2026
83.87Pass@1 Acc
8
Feb 26, 2026
50.25Accuracy
8
Feb 26, 2026
45Accuracy
8
Feb 26, 2026
44.06Accuracy
7
Jun 24, 2026
69.89Accuracy
7
Jun 4, 2026
55.7Accuracy
6
Jun 18, 2026