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

OPUS-MT Dashboard

Model Scores (selected model)

IDLanguageBenchmarkOutputbleu
0eng-belflores101-devtestshow 11.2
1eng-belflores200-devtestshow 11.2
2eng-belntrex128show (logfile)22.8
3eng-beltatoeba-test-v2020-07-28show 24.9
4eng-beltatoeba-test-v2021-03-30show 24.9
5eng-beltatoeba-test-v2021-08-07show 24.9
6eng-rusflores101-devtestshow 32.7
7eng-rusflores200-devtestshow 32.7
8eng-rusgeneraltest2022show (logfile)31.1
9eng-rusnewstest2012show 36.8
10eng-rusnewstest2013show 26.9
11eng-rusnewstest2014show 43.5
12eng-rusnewstest2015show 34.9
13eng-rusnewstest2016show 33.1
14eng-rusnewstest2017show 37.3
15eng-rusnewstest2018show 32.9
16eng-rusnewstest2019show 31.8
17eng-rusnewstest2020show 25.5
18eng-rusnewstest2021show (logfile)39.8
19eng-rusntrex128show (logfile)34.6
20eng-rustatoeba-test-v2020-07-28show 46.3
21eng-rustatoeba-test-v2021-03-30show 45.2
22eng-rustatoeba-test-v2021-08-07show 45.5
23eng-rustico19-testshow 33.7
24eng-ukrflores101-devtestshow 32.1
25eng-ukrflores200-devtestshow 32.1
26eng-ukrgeneraltest2022show (logfile)25.9
27eng-ukrntrex128show (logfile)26.7
28eng-ukrtatoeba-test-v2020-07-28show 37.4
29eng-ukrtatoeba-test-v2021-03-30show 37.7
30eng-ukrtatoeba-test-v2021-08-07show 37.7
average32.058

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