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

OPUS-MT Dashboard

Model Scores (selected model)

IDLanguageBenchmarkOutputbleu
0ces-engflores101-devtestshow 29.4
1ces-engflores200-devtestshow 29.3
2ces-enggeneraltest2022show (logfile)39.1
3ces-engmulti30k_test_2016_flickrshow 28.9
4ces-engmulti30k_test_2018_flickrshow 27.2
5ces-engnewssyscomb2009show 23.4
6ces-engnewstest2008show (logfile)20.7
7ces-engnewstest2009show 22.6
8ces-engnewstest2010show 22.8
9ces-engnewstest2011show 23.8
10ces-engnewstest2012show 23.1
11ces-engnewstest2013show 25.8
12ces-engnewstest2014show 27.2
13ces-engnewstest2015show 25.2
14ces-engnewstest2016show 26.7
15ces-engnewstest2017show 23.7
16ces-engnewstest2018show 24.6
17ces-engnewstest2021show (logfile)30.7
18ces-engntrex128show (logfile)25.9
19ces-engtatoeba-test-v2020-07-28show 50.6
20ces-engtatoeba-test-v2021-03-30show 50.1
21ces-engtatoeba-test-v2021-08-07show 49.8
22pol-engflores101-devtestshow 20.4
23pol-engflores200-devtestshow 20.3
24pol-engnewstest2020show 24.9
25pol-engntrex128show (logfile)20.9
26pol-engtatoeba-test-v2020-07-28show 48.1
27pol-engtatoeba-test-v2021-03-30show 48.1
28pol-engtatoeba-test-v2021-08-07show 48.1
average30.393

GitHub: [OPUS] [OPUS-MT] [benchmarks] [leaderboards] [dashboard]