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

OPUS-MT Dashboard

Model Scores (selected model)

IDLanguageBenchmarkOutputbleu
0est-engflores101-devtestshow 25.8
1est-engflores200-devtestshow 25.8
2est-engnewstest2018show 23.7
3est-engntrex128show 22.6
4est-engtatoeba-test-v2020-07-28show 52.7
5est-engtatoeba-test-v2021-03-30show 52.7
6est-engtatoeba-test-v2021-08-07show 52.7
7fin-engflores101-devtestshow 23.8
8fin-engflores200-devtestshow 23.9
9fin-engnewstest2015show 21.5
10fin-engnewstest2016show 23.3
11fin-engnewstest2017show 25.6
12fin-engnewstest2018show 18.9
13fin-engnewstest2019show 23.2
14fin-engnewstestB2016show 19.3
15fin-engnewstestB2017show 22
16fin-engntrex128show 20
17fin-engtatoeba-test-v2020-07-28show 46.5
18fin-engtatoeba-test-v2021-03-30show 46.4
19fin-engtatoeba-test-v2021-08-07show 46.1
20hun-engflores101-devtestshow 24.6
21hun-engflores200-devtestshow 24.6
22hun-engnewssyscomb2009show 18.3
23hun-engnewstest2008show 16.1
24hun-engnewstest2009show 17.3
25hun-engntrex128show 16.4
26hun-engtatoeba-test-v2020-07-28show 45
27hun-engtatoeba-test-v2021-03-30show 43.1
28hun-engtatoeba-test-v2021-08-07show 42.4
29liv-enggeneraltest2022show 3.9
average28.940

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