[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 26.2
1est-engflores200-devtestshow 26.2
2est-engnewstest2018show 24
3est-engntrex128show (logfile)23.1
4est-engtatoeba-test-v2020-07-28show 52.8
5est-engtatoeba-test-v2021-03-30show 52.8
6est-engtatoeba-test-v2021-08-07show 52.8
7fin-engflores101-devtestshow 24
8fin-engflores200-devtestshow 23.9
9fin-engnewstest2015show 21.9
10fin-engnewstest2016show 23.1
11fin-engnewstest2017show 25.8
12fin-engnewstest2018show 19.3
13fin-engnewstest2019show 22.8
14fin-engnewstestB2016show 19.3
15fin-engnewstestB2017show 22.2
16fin-engntrex128show (logfile)20.3
17fin-engtatoeba-test-v2020-07-28show 46.4
18fin-engtatoeba-test-v2021-03-30show 46.2
19fin-engtatoeba-test-v2021-08-07show 46
20hun-engflores101-devtestshow 24.4
21hun-engflores200-devtestshow 24.5
22hun-engnewssyscomb2009show 19.2
23hun-engnewstest2008show (logfile)16.3
24hun-engnewstest2009show 17.9
25hun-engntrex128show (logfile)16.4
26hun-engtatoeba-test-v2020-07-28show 44.3
27hun-engtatoeba-test-v2021-03-30show 42.5
28hun-engtatoeba-test-v2021-08-07show 42
29liv-enggeneraltest2022show (logfile)4.1
average29.023

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