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

LlamaQ

Benchmarks

Task NameDataset NameSOTA ResultTrend
Speech-to-Text Question-AnsweringLlamaQ
Accuracy86
26
Factuality EvaluationLlamaQ
Response Accuracy88.4
18
Speech-to-Speech Question-AnsweringLlamaQ
Accuracy76.5
17
Spoken Question AnsweringLlamaQ S→T
Accuracy75.7
10
Spoken Question AnsweringLLaMAQ
Text Score74.2
6
Showing 5 of 5 rows