[restart] [share link] select language: [swap] [compare scores] [compare models] [map] [release history] [uploads]

OPUS-MT Dashboard

Model Scores (comparing between OPUS-MT and external models)

IDBenchmark (bleu)OutputOPUS-MTbleuexternalbleuDiff
0flores101-devtestcomparefra-eng/opus..2022-03-0946facebook/nllb-200-3.3B45.4 0.6
1flores200-devtestcomparefra-eng/opus..2022-03-0946facebook/nllb-200-3.3B45.7 0.3
2multi30k_test_2016_flickrcompareroa-eng/opus..2024-08-1749.8facebook/nll..illed-1.3B49.5 0.3
3multi30k_test_2017_flickrcompareroa-eng/opus..2024-08-1752.2facebook/nllb-200-1.3B50.7 1.5
4multi30k_test_2017_mscococompareroa-eng/opus..2024-08-1751.9facebook/nll..illed-1.3B50.9 1.0
5multi30k_test_2018_flickrcompareroa-eng/opus..2024-08-1745.4facebook/nllb-200-1.3B44.9 0.5
6newsdiscusstest2015compareroa-eng/opus..2024-08-1740.7facebook/nllb-200-3.3B36.6 4.1
7newssyscomb2009compareroa-eng/opus..2024-08-1731.7facebook/nllb-200-1.3B30.7 1.0
8newstest2008compareroa-deu+eng+..2024-05-3026.7facebook/nllb-200-3.3B26.9-0.2
9newstest2009compareroa-deu+eng+..2024-05-3030.6facebook/nllb-200-1.3B30.1 0.5
10newstest2010comparefra-eng/opus..2022-03-0933.4facebook/nllb-200-1.3B34.1-0.7
11newstest2011compareroa-deu+eng+..2024-05-3033.9facebook/nllb-200-3.3B33.8 0.1
12newstest2012compareroa-deu+eng+..2024-05-3033.7facebook/nllb-200-3.3B33.5 0.2
13newstest2013comparefra-eng/opus..2022-03-0934.8facebook/nllb-200-1.3B34.7 0.1
14newstest2014compareroa-eng/opus..2024-08-1739.7facebook/nllb-200-3.3B40.3-0.6
15ntrex128compareroa-eng/opus..2024-08-1737.5facebook/nllb-200-3.3B39-1.5
16tatoeba-test-v2020-07-28compareroa-eng/opus..2024-08-1759.6facebook/nllb-200-3.3B60.2-0.6
17tatoeba-test-v2021-03-30compareroa-eng/opus..2024-08-1759.8facebook/nllb-200-3.3B60.6-0.8
18tatoeba-test-v2021-08-07compareroa-eng/opus..2024-08-1760facebook/nllb-200-3.3B60 0.0
19tico19-testcompareroa-eng/opus..2024-08-1741.6facebook/nllb-200-3.3B43.4-1.8
average42.842.6 0.2

GitHub: [OPUS] [OPUS-MT] [benchmarks] [leaderboards] [dashboard]