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

HelloBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Open-ended generationHelloBench (HB)
HB-A Score84
17
Writing EvaluationHelloBench-G
Score62.89
12
Writing EvaluationHelloBench-C
Score52.98
12
Open-ended writingHelloBench
Average Score82
11
Creative WritingHelloBench
Accuracy80.4
6
Heuristic Text GenerationHelloBench HTG
ROUGE-L (2K context)0.201
5
Showing 6 of 6 rows