[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 40.6
1deu-engmulti30k_task2_test_2016show 4
2deu-engmulti30k_test_2016_flickrshow 39.7
3deu-engmulti30k_test_2017_flickrshow 40.3
4deu-engmulti30k_test_2017_mscocoshow 32.5
5deu-engmulti30k_test_2018_flickrshow 36.4
6deu-engnewssyscomb2009show 29.3
7deu-engnewstest2009show 26.8
8deu-engnewstest2010show 30
9deu-engnewstest2011show 27.3
10deu-engnewstest2012show 29
11deu-engnewstest2013show 32
12deu-engnewstest2014show 34
13deu-engnewstest2015show 34.1
14deu-engnewstest2016show 40.2
15deu-engnewstest2017show 35.6
16deu-engnewstest2018show 43.5
17deu-engnewstest2019show 39.3
18deu-engnewstest2020show 39.5
19deu-engnewstestB2020show 38.6
20deu-engtatoeba-test-v2020-07-28show 55.7
21deu-engtatoeba-test-v2021-03-30show 54.2
22deu-engtatoeba-test-v2021-08-07show 52.7
average36.317

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