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

INCLUDE

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multilingual Language ProficiencyINCLUDE base-44
Average Score64.8
46
Language UnderstandingINCLUDE base 44
Average Score64.8
21
Multilingual Multiple-Choice ReasoningINCLUDE 44 languages 1.0 (test)
Average Accuracy56.9
6
Isolated Sign Language RecognitionINCLUDE
Accuracy93.5
5
Multilingual KnowledgeINCLUDE
Accuracy77.2
4
Multilingual Language UnderstandingINCLUDE 5-shot
Accuracy77.81
3
Showing 6 of 6 rows