Problem-Solving
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
203Accuracy
24
Apr 15, 2026
78.94Exact Match Accuracy
20
Mar 19, 2026
81.97MMLU Problem Solving Accuracy
19
May 14, 2026
60.12Exact Match (%)
15
Mar 16, 2026
22.2End Pass@1
6
Jun 23, 2026