Scientific Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
65.8Score
14
Mar 27, 2026
69Score
14
Mar 27, 2026
59.48Success Rate (SR)
14
Feb 26, 2026
66Initial Score
13
May 20, 2026
33.1Overall Accuracy
13
Feb 26, 2026
8Samples Required
12
Jul 1, 2026
25.96Mean Score @16
12
Jun 23, 2026
27.28Mean@16
12
Jun 23, 2026
5,318Thinking Tokens
12
Jun 17, 2026
36.98Wall-clock Time (s)
12
Jun 11, 2026
48.1pass@1
12
Jun 4, 2026
98pass@k
12
May 28, 2026
64Accuracy
12
May 6, 2026
66.2Accuracy
12
Apr 29, 2026
29.64Pass@1
12
Jun 2, 2026
64.96Accuracy
11
Jun 26, 2026
2,004.7Token Count
11
Apr 24, 2026
20.8Accuracy
11
Apr 8, 2026
26.6AR
11
Feb 26, 2026
69.7Score
11
Feb 26, 2026
67.2Pass@1
11
Mar 16, 2026
87.37Average Score
11
Feb 26, 2026
43Accuracy
10
Jul 7, 2026
81.6Accuracy
10
Jun 30, 2026
81.5Accuracy
10
Jun 30, 2026