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

OpenQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Open Question AnsweringOpenQA
Accuracy88.2
8
Commonsense ReasoningOpenQA (eval)
Accuracy75.4
8
Open-domain Question AnsweringOpenQA
Output Token Throughput (tok/s)58.47
2
Showing 3 of 3 rows