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

CharacterEval

Benchmarks

Task NameDataset NameSOTA ResultTrend
Character ConsistencyCharacterEval
KE2.58
25
Character Role-playingCharacterEval
Coherence4.14
18
Role-playing AttractivenessCharacterEval
HL3.618
13
Conversational AbilityCharacterEval
Fluency3.612
13
general character taskCharacterEval
Win Rate65.3
8
Character EvaluationCharacterEval
Score3.23
7
Role-playingCharacterEval unseen roles transfer setting
CC2.749
4
Showing 7 of 7 rows