Instruction Following
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
7.76MT-Bench Score
12
Feb 26, 2026
84.53Accuracy
11
Jul 7, 2026
79.4LC Score
11
May 21, 2026
73.4Accuracy
11
Jul 7, 2026
0.9NMSE
11
May 4, 2026
20.66Success Rate (SR)
11
Apr 28, 2026
79.19Success Rate (SR)
11
Apr 28, 2026
79.67MMLU
11
Apr 21, 2026
16.31ROUGE-L
11
Mar 4, 2026
86.8POPE
11
Feb 26, 2026
62.2Success Rate (SR)
11
Feb 26, 2026
14Win Rate
11
Feb 26, 2026
7.22OLMo-2 Reward Score
10
Jun 4, 2026
99Clean Rate
10
May 12, 2026
50.47Accuracy
10
Apr 30, 2026
69.66Accuracy
10
Apr 30, 2026
17.4Accuracy
10
Apr 24, 2026
9.77Internal Score
10
Mar 18, 2026
93.16IFEval Accuracy
10
Mar 18, 2026
61.4Instruction Level Strict Accuracy
10
Mar 4, 2026
31.3Win Rate (Raw)
10
Feb 26, 2026
0.203LC
10
Feb 26, 2026
95.4PDD Instruction Following Score
10
Feb 26, 2026
100Punctuation Score
10
Feb 26, 2026
22.9AlpacaEval2 Score
10
Feb 26, 2026