[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 (chrf)OutputOPUS-MTchrfexternalchrfDiff
0flores101-devtestcomparefra-eng/opus..2022-03-090.6935facebook/nllb-200-3.3B0.687580.006
1flores200-devtestcomparefra-eng/opus..2022-03-090.6935facebook/nllb-200-3.3B0.68970.004
2multi30k_test_2016_flickrcomparefra-eng/opus..2022-03-090.68005facebook/nll..illed-1.3B0.68136-0.001
3multi30k_test_2017_flickrcomparefra-eng/opus..2022-03-090.70596facebook/nllb-200-1.3B0.691040.015
4multi30k_test_2017_mscococomparefra-eng/opus..2022-03-090.69356facebook/nllb-200-3.3B0.69966-0.006
5multi30k_test_2018_flickrcomparefra-eng/opus..2022-03-090.65751facebook/nllb-200-3.3B0.652640.005
6newsdiscusstest2015comparefra-eng/opus..2022-03-090.62603facebook/nllb-200-3.3B0.599090.027
7newssyscomb2009comparefra-eng/opus..2022-03-090.57488facebook/nllb-200-1.3B0.568470.006
8newstest2008comparefra-eng/opus..2022-03-090.5431facebook/nllb-200-1.3B0.542950.000
9newstest2009comparefra-eng/opus-2021-02-190.57119facebook/nllb-200-3.3B0.564120.007
10newstest2010comparefra-eng/opus..2022-03-090.59561facebook/nllb-200-1.3B0.59709-0.001
11newstest2011comparefra-eng/opus..2022-03-090.60271facebook/nllb-200-3.3B0.597320.005
12newstest2012comparefra-eng/opus..2022-03-090.59507facebook/nllb-200-1.3B0.593230.002
13newstest2013comparefra-eng/opus..2022-03-090.59691facebook/nllb-200-1.3B0.591080.006
14newstest2014comparefra-eng/opus..2022-03-090.64533facebook/nllb-200-3.3B0.645110.000
15tatoeba-test-v2020-07-28comparefra-eng/opus..2022-03-090.73088facebook/nllb-200-3.3B0.73776-0.007
16tatoeba-test-v2021-03-30comparefra-eng/opus..2022-03-090.73444facebook/nllb-200-3.3B0.74179-0.007
17tatoeba-test-v2021-08-07comparefra-eng/opus..2022-03-090.73772facebook/nllb-200-3.3B0.73885-0.001
18tico19-testcomparefra-eng/opus..2022-03-090.63326facebook/nllb-200-3.3B0.65213-0.019
average0.6480.6460.002