Instruction Following
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
95IFEval Accuracy
854
Jun 26, 2026
95.87Win Rate
752
Jun 12, 2026
98.4Win Rate
423
Jul 9, 2026
9.32MT-Bench Score
287
Jun 29, 2026
98.11Win Rate
263
Jun 2, 2026
75.27Dolly (BLEU)
224
Mar 26, 2026
40.2Rouge-L
178
May 21, 2026
5.27Speedup (x)
173
Jun 29, 2026
75.4LC (%)
147
Jul 7, 2026
38.7Rouge-L
119
Mar 24, 2026
32.5Rouge-L
114
Mar 5, 2026
71Accuracy
102
Jul 7, 2026
90.39Accuracy (IFEval)
101
Jun 4, 2026
20.93Rouge-L
101
May 21, 2026
67.45LC Win Rate (%)
95
May 19, 2026
94.64IFEval Score
92
Jul 7, 2026
84.8IFEval Score
92
Jun 23, 2026
97.9Rouge-L
90
Feb 26, 2026
79HSR
85
May 29, 2026
89.3Accuracy
82
Feb 26, 2026
21.85Agriculture Score
80
Mar 26, 2026
40.75VicunaEval Score
80
Feb 27, 2026
21.7Rouge-L
73
Apr 6, 2026
77.8Accuracy
72
Apr 29, 2026
82.9Pass@1 (Strict)
72
Mar 20, 2026