Scientific Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95.9Accuracy
164
Feb 26, 2026
92.5Accuracy
121
Jun 17, 2026
67.97Accuracy
115
Jun 5, 2026
74.24Accuracy
94
Jul 9, 2026
99.37Accuracy
88
Jun 18, 2026
96.21Score
86
Jun 26, 2026
66.2Accuracy (%)
77
May 12, 2026
70.2Accuracy
75
Mar 31, 2026
80.13Accuracy
73
Jun 26, 2026
82.49Accuracy
68
May 1, 2026
73.2Score
68
Apr 10, 2026
78.66Pass@1 Accuracy
67
Jun 2, 2026
82.38Chemistry Accuracy
56
Jul 3, 2026
83.8Accuracy
55
Feb 26, 2026
65.25Accuracy
54
Jun 4, 2026
4.43Latency
54
May 12, 2026
48.74RealMath Score
50
Mar 20, 2026
87.5Accuracy
48
Apr 3, 2026
70.2pass@1
43
May 15, 2026
87.4Accuracy
43
Jun 17, 2026
25.2IPhO 2025 Score
43
Feb 26, 2026
0.382Accuracy
42
Jul 3, 2026
1Total Inference Runtime (s)
36
Apr 6, 2026
82.8Accuracy
36
Mar 17, 2026
79.09Math Score
36
May 19, 2026