Instruction Following
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
97.35Accuracy
4
Jun 23, 2026
86.98Accuracy
4
Jun 23, 2026
98.16Accuracy
4
Jun 23, 2026
93.16Accuracy
4
Jun 23, 2026
2.4LC-WR (Win Rate)
4
Jun 17, 2026
57.6Score
4
Jun 16, 2026
72.2Loose Prompt Success Rate
4
Jun 9, 2026
30.33Wizard Score
4
May 26, 2026
60.19Instruction Following Score
4
May 13, 2026
49.72Instruction Following Rate
4
May 13, 2026
72.79F1 Score
4
May 13, 2026
56.19Pass@4
4
May 12, 2026
80Success Rate
4
Apr 28, 2026
73.8Prompt Loose Score
4
Apr 21, 2026
28.4ROUGE-1 Score
4
Mar 25, 2026
84.5Worst@20 Score
4
Mar 23, 2026
58.3Pass Rate
4
Mar 20, 2026
0.554Comparison Score (CS)
4
Mar 18, 2026
52Win Rate
4
Mar 6, 2026
80.8Win Rate
4
Mar 5, 2026
90.9Average Win Rate
4
Mar 5, 2026
0.445Accuracy
4
Mar 4, 2026
36Distinct-1
4
Mar 4, 2026
74.12Accuracy
4
May 27, 2026
60.8Accuracy
4
Feb 26, 2026