Expert-Level Reasoning
Benchmarks
DatasetSOTA methodMetricTrendResultsLast updated
84.3Pass@1 Score
20
Jul 7, 2026
54.2Avg@3 Score
15
Mar 18, 2026
52.2Inference Accuracy
13
Feb 26, 2026
0.9Inference Accuracy
12
Feb 26, 2026
81.6Inference Accuracy
12
Feb 26, 2026
26.5Accuracy (HLE)
5
Jul 7, 2026