IFEval
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
IFEval MultiIF
90.65I-Following Score
20
IFEval
85.1Accuracy
20
IFEval
96.3IFEval Score
18
IFEval
61.6Accuracy
18
IFEval Inverse
83.7Accuracy
18
IFEval loose
78.1Avg@8
18
IFEval
88.36Accuracy
17
IFEval
91.31Accuracy
17
IFEval
89.8Score
17
IFEval
84.1Precision (Short)
16
IFEval
46.2Avg@32 Accuracy
16
IFEval
82.4Pass@1 (Strict)
16
IFEval
54.4Average Score
16
IFEval
84.3IFEval Score
15
IFEval
23Arabic IFEval Score
15
IFEval
71.2Strict Accuracy
15
IFEval
90.77IFEval Score
15
IFEval Alpaca
60.31Compliance Score
15
IFEval Dolly-15K
58.71Compliance Score
15
IFEval
48.8Loose Prompt Metric
15
IFEVAL
75.7SSR
14
IFEval-PT
79.33Instruction Score
14
IFEval
95.8Normalized performance
13
IFEval Spanish (ES)
82.92Strict Accuracy
13
IFEval Catalan (CA)
76.81Strict Accuracy
13