[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 25.1
1est-engflores200-devtestshow 24.9
2est-engnewstest2018show 22.5
3est-engntrex128show (logfile)21.3
4est-engtatoeba-test-v2020-07-28show 52.1
5est-engtatoeba-test-v2021-03-30show 52.1
6est-engtatoeba-test-v2021-08-07show 52.1
7fin-engflores101-devtestshow 22.9
8fin-engflores200-devtestshow 22.9
9fin-engnewstest2015show 20.3
10fin-engnewstest2016show 21.6
11fin-engnewstest2017show 24.5
12fin-engnewstest2018show 18.5
13fin-engnewstest2019show 21.2
14fin-engnewstestB2016show 18
15fin-engnewstestB2017show 20.8
16fin-engntrex128show (logfile)19
17fin-engtatoeba-test-v2020-07-28show 45.4
18fin-engtatoeba-test-v2021-03-30show 45.2
19fin-engtatoeba-test-v2021-08-07show 45
20hun-engflores101-devtestshow 23.2
21hun-engflores200-devtestshow 23.6
22hun-engnewssyscomb2009show 18.5
23hun-engnewstest2008show (logfile)15.6
24hun-engnewstest2009show 17.8
25hun-engntrex128show (logfile)15.2
26hun-engtatoeba-test-v2020-07-28show 42.8
27hun-engtatoeba-test-v2021-03-30show 41
28hun-engtatoeba-test-v2021-08-07show 40.4
29liv-enggeneraltest2022show (logfile)3.7
average27.907

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