Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

JSON-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Formal Language GenerationJSON-Bench
Syntactic Accuracy @199.7
16
Constrained DecodingJSON-Bench
Latency (s)8.15
16
Functional CorrectnessJSON-Bench
Functional Accuracy @155.2
16
Showing 3 of 3 rows