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

Internal Evaluation

Benchmarks

Task NameDataset NameSOTA ResultTrend
Subjective Search Question AnsweringInternal Evaluation Subjective Q&A
V4 Win Count190
16
Objective Search Question AnsweringInternal Evaluation Objective Q&A
V4 Wins269
10
Multi-Turn WritingInternal Evaluation
Tie Score4
2
Complex Instruction FollowingInternal Evaluation
Tie Score0
2
Showing 4 of 4 rows