[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 40.8
1deu-engflores200-devtestshow 40.8
2deu-enggeneraltest2022show (logfile)47
3deu-engmulti30k_task2_test_2016show 4
4deu-engmulti30k_test_2016_flickrshow 40
5deu-engmulti30k_test_2017_flickrshow 41.1
6deu-engmulti30k_test_2017_mscocoshow 33.9
7deu-engmulti30k_test_2018_flickrshow 36.7
8deu-engnewssyscomb2009show 29.1
9deu-engnewstest2008show (logfile)27.6
10deu-engnewstest2009show 27.1
11deu-engnewstest2010show 30.5
12deu-engnewstest2011show 27.6
13deu-engnewstest2012show 29.3
14deu-engnewstest2013show 32.3
15deu-engnewstest2014show 34.2
16deu-engnewstest2015show 34.2
17deu-engnewstest2016show 40.4
18deu-engnewstest2017show 36.1
19deu-engnewstest2018show 44.3
20deu-engnewstest2019show 39.7
21deu-engnewstest2020show 39.9
22deu-engnewstest2021show (logfile)46.9
23deu-engnewstestALL2020show (logfile)53.4
24deu-engnewstestB2020show 38.5
25deu-engntrex128show (logfile)34.4
26deu-engtatoeba-test-v2020-07-28show 53.5
27deu-engtatoeba-test-v2021-03-30show 52.2
28deu-engtatoeba-test-v2021-08-07show 50.8
average37.459