Expert-Level Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
66.65Pass@1
39
Jun 1, 2026
61.7Best Accuracy
25
May 12, 2026
40.9Accuracy
14
Feb 26, 2026
68.9Accuracy @8
6
Jun 5, 2026
40.2Accuracy
5
Jun 1, 2026
17.27LLMcritic Calls
5
May 11, 2026
34.51GPQA Score
2
Feb 26, 2026