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

World Knowledge

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multiple-Choice Question AnsweringWorld Knowledge Average of OBQA, ARC-C, ARC-E, SCIQ, SIQA
Average Accuracy87.1
66
Agentic RoutingWorld Knowledge (WK)
Accuracy38
10
Showing 2 of 2 rows