[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 25.3
1est-engflores200-devtestshow 25.3
2est-engnewstest2018show 22.8
3est-engntrex128show (logfile)21.7
4est-engtatoeba-test-v2020-07-28show 52.2
5est-engtatoeba-test-v2021-03-30show 52.2
6est-engtatoeba-test-v2021-08-07show 52.2
7fin-engflores101-devtestshow 23
8fin-engflores200-devtestshow 23.1
9fin-engnewstest2015show 21
10fin-engnewstest2016show 22.6
11fin-engnewstest2017show 25
12fin-engnewstest2018show 18.3
13fin-engnewstest2019show 22.9
14fin-engnewstestB2016show 18.7
15fin-engnewstestB2017show 21.4
16fin-engntrex128show (logfile)19.1
17fin-engtatoeba-test-v2020-07-28show 45.1
18fin-engtatoeba-test-v2021-03-30show 44.9
19fin-engtatoeba-test-v2021-08-07show 44.8
20hun-engflores101-devtestshow 23.9
21hun-engflores200-devtestshow 23.8
22hun-engnewssyscomb2009show 18.6
23hun-engnewstest2008show (logfile)15.5
24hun-engnewstest2009show 17.3
25hun-engntrex128show (logfile)15.6
26hun-engtatoeba-test-v2020-07-28show 43.8
27hun-engtatoeba-test-v2021-03-30show 41.9
28hun-engtatoeba-test-v2021-08-07show 41.2
29liv-enggeneraltest2022show (logfile)4
average28.240

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