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

CortexBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Question AnsweringCORTEXBENCH Comparison
Accuracy95.7
32
Question AnsweringCORTEXBENCH Bridge
Accuracy96.1
32
Robot LearningCortexBench
Adroit Score74.7
22
Evidence RetrievalCORTEXBENCH Comparison
Hit@546
8
Evidence RetrievalCORTEXBENCH Bridge
Hit@538.4
8
Object NavigationCortexBench ObjNav
Success80
3
Showing 6 of 6 rows