[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 41.2
1deu-engflores200-devtestshow 41.4
2deu-enggeneraltest2022show (logfile)46.5
3deu-engmulti30k_task2_test_2016show 4.0
4deu-engmulti30k_test_2016_flickrshow 40.0
5deu-engmulti30k_test_2017_flickrshow 40.9
6deu-engmulti30k_test_2017_mscocoshow 33.0
7deu-engmulti30k_test_2018_flickrshow 36.9
8deu-engnewssyscomb2009show 29.2
9deu-engnewstest2008show (logfile)27.7
10deu-engnewstest2009show 26.7
11deu-engnewstest2010show 30.4
12deu-engnewstest2011show 27.2
13deu-engnewstest2012show 29.1
14deu-engnewstest2013show 32.2
15deu-engnewstest2014show 34.0
16deu-engnewstest2015show 34.3
17deu-engnewstest2016show 40.8
18deu-engnewstest2017show 36.0
19deu-engnewstest2018show 44.4
20deu-engnewstest2019show 39.5
21deu-engnewstest2020show 39.5
22deu-engnewstest2021show (logfile)45.9
23deu-engnewstestALL2020show (logfile)53.0
24deu-engnewstestB2020show 38.2
25deu-engntrex128show (logfile)34.5
26deu-engtatoeba-test-v2020-07-28show 53.3
27deu-engtatoeba-test-v2021-03-30show 51.9
28deu-engtatoeba-test-v2021-08-07show 50.5
average37.317