| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | est-eng | flores101-devtest | show | 31 |
| 1 | est-eng | newstest2018 | show | 28 |
| 2 | est-eng | tatoeba-test-v2021-08-07 | show | 57.7 |
| 3 | fin-eng | flores101-devtest | show | 28.6 |
| 4 | fin-eng | newstest2015 | show | 25.6 |
| 5 | fin-eng | newstest2016 | show | 27.6 |
| 6 | fin-eng | newstest2017 | show | 30.7 |
| 7 | fin-eng | newstest2018 | show | 23.5 |
| 8 | fin-eng | newstest2019 | show | 27.2 |
| 9 | fin-eng | newstestB2016 | show | 23.2 |
| 10 | fin-eng | newstestB2017 | show | 25.7 |
| 11 | fin-eng | tatoeba-test-v2020-07-28 | show | 53 |
| 12 | fin-eng | tatoeba-test-v2021-03-30 | show | 53 |
| 13 | fin-eng | tatoeba-test-v2021-08-07 | show | 52.8 |
| average | 34.829 |