Scientific Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
55.5Score
5
Mar 27, 2026
30.34Accuracy
5
Feb 26, 2026
54.64Accuracy
5
Feb 26, 2026
80.22Average Score @8
4
Jun 25, 2026
56.79Avg@8
4
Jun 25, 2026
71.86Average Score @8
4
Jun 25, 2026
61.9Physics Accuracy
4
Jun 16, 2026
59.1Pass@1 Accuracy
4
Jun 4, 2026
65.2Accuracy
4
May 22, 2026
65.38Biology Accuracy
4
May 15, 2026
34.7Avg@4
4
May 12, 2026
78.3Accuracy
4
Apr 15, 2026
27Accuracy
4
Apr 10, 2026
70.65Accuracy
4
Mar 18, 2026
55.7Pass@1
4
Feb 26, 2026
93.8Best@16
3
Jun 30, 2026
87.9Best@16
3
Jun 30, 2026
80.1Best@16
3
Jun 30, 2026
26.3Accuracy (GPQA Hard)
3
Jun 12, 2026
16.82Score
3
May 29, 2026
65.6SQA-I Score
3
May 26, 2026
74.49Accuracy
3
Apr 21, 2026
66.4Scientific Reasoning QA Accuracy
3
Feb 26, 2026
60.4% Avg@4
3
Feb 26, 2026
0.405Average Accuracy
3
Feb 26, 2026