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

OPUS-MT Dashboard

Model Scores (selected model)

IDLanguageBenchmarkOutputbleu
0bos_Latn-engtatoeba-test-v2020-07-28show 66.7
1bos_Latn-engtatoeba-test-v2021-03-30show 66.7
2bos_Latn-engtatoeba-test-v2021-08-07show 66.5
3hbs-engtatoeba-test-v2020-07-28show 56.4
4hbs-engtatoeba-test-v2021-03-30show 56.4
5hbs-engtatoeba-test-v2021-08-07show 56.4
6hrv-engflores101-devtestshow 37.1
7hrv-engflores200-devtestshow 37.1
8hrv-engntrex128show 38
9hrv-engtatoeba-test-v2020-07-28show 58.7
10hrv-engtatoeba-test-v2021-03-30show 58.7
11hrv-engtatoeba-test-v2021-08-07show 58.8
12srp_Cyrl-engflores101-devtestshow 35.9
13srp_Cyrl-engflores200-devtestshow 35.9
14srp_Cyrl-engntrex128show 27.2
15srp_Cyrl-engtatoeba-test-v2020-07-28show 44.7
16srp_Cyrl-engtatoeba-test-v2021-03-30show 44.7
17srp_Cyrl-engtatoeba-test-v2021-08-07show 44.7
18srp_Latn-engntrex128show 30.5
19srp_Latn-engtatoeba-test-v2020-07-28show 58.4
20srp_Latn-engtatoeba-test-v2021-03-30show 58.4
21srp_Latn-engtatoeba-test-v2021-08-07show 58.4
average49.832

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