[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 41
1deu-engflores200-devtestshow 41
2deu-enggeneraltest2022show (logfile)47.3
3deu-engmulti30k_task2_test_2016show 4.1
4deu-engmulti30k_test_2016_flickrshow 40.5
5deu-engmulti30k_test_2017_flickrshow 41.5
6deu-engmulti30k_test_2017_mscocoshow 33.3
7deu-engmulti30k_test_2018_flickrshow 37.2
8deu-engnewssyscomb2009show 28.4
9deu-engnewstest2008show (logfile)27.9
10deu-engnewstest2009show 27.1
11deu-engnewstest2010show 30.8
12deu-engnewstest2011show 27.4
13deu-engnewstest2012show 29.4
14deu-engnewstest2013show 32.4
15deu-engnewstest2014show 34.4
16deu-engnewstest2015show 34.6
17deu-engnewstest2016show 40.7
18deu-engnewstest2017show 36.3
19deu-engnewstest2018show 44.4
20deu-engnewstest2019show 39.9
21deu-engnewstest2020show 40.1
22deu-engnewstest2021show (logfile)47.5
23deu-engnewstestALL2020show (logfile)53.6
24deu-engnewstestB2020show 38.7
25deu-engntrex128show (logfile)34.4
26deu-engtatoeba-test-v2020-07-28show 53.6
27deu-engtatoeba-test-v2021-03-30show 52.4
28deu-engtatoeba-test-v2021-08-07show 51.1
average37.621