Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

OPUS-100

Benchmarks

Task NameDataset NameSOTA ResultTrend
Machine TranslationOPUS-100 (test)
Average BLEU Score39.1
19
Unsupervised Machine Translationopus-100 Pt->En 1.0 (test)
Tokenized BLEU30.5
3
Unsupervised Machine Translationopus-100 En->Pt 1.0 (test)
Tokenized BLEU18.4
3
Multilingual Machine TranslationOPUS-100 Supervised x-en directions
BLEU Score20.1
2
Multilingual Machine TranslationOPUS-100 Zero-shot directions (x-y)
OTR72
2
Machine TranslationOPUS-100 v1.0 (test)
MT Score (En->Be)-
0
Showing 6 of 6 rows