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

OPUS-MT Dashboard

Model Scores (selected model)

IDLanguageBenchmarkOutputbleu
0deu-engflores101-devtestshow 37.3
1deu-engmulti30k_task2_test_2016show 4
2deu-engmulti30k_test_2016_flickrshow 39
3deu-engmulti30k_test_2017_flickrshow 37.5
4deu-engmulti30k_test_2017_mscocoshow 29.9
5deu-engmulti30k_test_2018_flickrshow 34.6
6deu-engnewssyscomb2009show 28
7deu-engnewstest2009show 25.9
8deu-engnewstest2010show 28.4
9deu-engnewstest2011show 26.2
10deu-engnewstest2012show 27.5
11deu-engnewstest2013show 30.3
12deu-engnewstest2014show 31.9
13deu-engnewstest2015show 32.3
14deu-engnewstest2016show 37.3
15deu-engnewstest2017show 33.3
16deu-engnewstest2018show 41
17deu-engnewstest2019show 36.1
18deu-engnewstest2020show 35.8
19deu-engnewstestB2020show 34.4
20deu-engtatoeba-test-v2020-07-28show 51.5
21deu-engtatoeba-test-v2021-03-30show 50.2
22deu-engtatoeba-test-v2021-08-07show 49
average33.974