[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 27.2
1est-engflores200-devtestshow 27.1
2est-engnewstest2018show 24.4
3est-engntrex128show (logfile)23.4
4est-engtatoeba-test-v2020-07-28show 53.3
5est-engtatoeba-test-v2021-03-30show 53.3
6est-engtatoeba-test-v2021-08-07show 53.3
7fin-engflores101-devtestshow 24.6
8fin-engflores200-devtestshow 24.7
9fin-engnewstest2015show 21.7
10fin-engnewstest2016show 24
11fin-engnewstest2017show 25.9
12fin-engnewstest2018show 19
13fin-engnewstest2019show 24
14fin-engnewstestB2016show 19.7
15fin-engnewstestB2017show 22.5
16fin-engntrex128show (logfile)20.1
17fin-engtatoeba-test-v2020-07-28show 46
18fin-engtatoeba-test-v2021-03-30show 45.9
19fin-engtatoeba-test-v2021-08-07show 45.7
20hun-engflores101-devtestshow 25.7
21hun-engflores200-devtestshow 25.7
22hun-engnewssyscomb2009show 19.7
23hun-engnewstest2008show (logfile)16
24hun-engnewstest2009show 17.9
25hun-engntrex128show (logfile)16.8
26hun-engtatoeba-test-v2020-07-28show 44.6
27hun-engtatoeba-test-v2021-03-30show 42.6
28hun-engtatoeba-test-v2021-08-07show 41.9
29liv-enggeneraltest2022show (logfile)4.6
average29.377

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