Text Quality Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.44Fluency
18
Feb 26, 2026
6.375PPL
17
Jul 1, 2026
7.43AlpacaEval
15
Feb 26, 2026
96.48Distinct-2
12
Feb 26, 2026
11.11Perplexity (PPL1)
5
Feb 26, 2026
3.2Fluency
4
Mar 4, 2026