Scientific Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
73.1Accuracy
6
Jun 18, 2026
69.9Accuracy
6
Jun 18, 2026
90.8Accuracy
6
Jun 18, 2026
70.8Accuracy
6
Jun 18, 2026
79.61Chemistry Score
6
Jun 11, 2026
60.5Pass Ratio
6
Jun 9, 2026
73.4Accuracy (Avg@8)
6
Jun 5, 2026
63.1GPQA Diamond Score
6
May 27, 2026
40.4Accuracy
6
May 19, 2026
48Accuracy
6
May 18, 2026
55.6Avg@8
6
Apr 21, 2026
193.62TPS
6
Apr 9, 2026
39.6Avg@64
6
Apr 7, 2026
46.46Acc@8
6
Mar 20, 2026
-0.55FPR
6
Feb 26, 2026
13.7Accuracy
6
Feb 26, 2026
20.2Pass@1
6
Feb 26, 2026
56.8avg@3
6
Feb 26, 2026
82.2avg@3
6
Feb 26, 2026
72.23Pass@10
6
Feb 26, 2026
11.44Error Rate (%)
6
Feb 26, 2026
42.4Pass@1
5
May 29, 2026
21Pass@1
5
May 29, 2026
56.4Score
5
Jun 4, 2026
82.81Accuracy
5
Apr 30, 2026