Instruction Following
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
80.81CSR
17
Apr 21, 2026
1,911Elo Score
16
Jun 4, 2026
5.33MT-Bench Score
16
Jun 4, 2026
84.1Precision (Short)
16
Jun 1, 2026
88.76Accuracy
16
May 26, 2026
82.4Pass@1 (Strict)
16
Apr 3, 2026
70Accuracy
15
Jun 9, 2026
23Arabic IFEval Score
15
May 28, 2026
90.77IFEval Score
15
May 26, 2026
12.568Delta Preference Score (bo64)
15
May 12, 2026
84.5Accuracy
15
May 12, 2026
73.96Accuracy
15
May 11, 2026
88Accuracy
15
Mar 12, 2026
7.67MT-Bench Score (10)
15
Feb 26, 2026
64.1Textual Entailment ROUGE-L F1
15
Feb 26, 2026
71.4SSR
14
Jun 5, 2026
75.7SSR
14
Jun 5, 2026
27.3IFB Score
14
Jun 2, 2026
78.57Clean F1
14
May 13, 2026
19.42ROUGE-L
14
Mar 24, 2026
79.33Instruction Score
14
Mar 5, 2026
84.66Accuracy
14
May 27, 2026
78.3Score
14
Feb 26, 2026
72.56Success Rate
14
Feb 26, 2026
97.7AlpacaEval Score
14
Feb 26, 2026