| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | eng-fin | flores101-devtest | show | 27.6 |
| 1 | eng-fin | flores200-devtest | show | 27.6 |
| 2 | eng-fin | newstest2015 | show | 26.4 |
| 3 | eng-fin | newstest2016 | show | 28.8 |
| 4 | eng-fin | newstest2017 | show | 31.3 |
| 5 | eng-fin | newstest2018 | show | 19.7 |
| 6 | eng-fin | newstest2019 | show | 26.4 |
| 7 | eng-fin | newstestALL2016 | show (logfile) | 32.7 |
| 8 | eng-fin | newstestALL2017 | show (logfile) | 35.8 |
| 9 | eng-fin | newstestB2016 | show | 23.3 |
| 10 | eng-fin | newstestB2017 | show | 26.8 |
| 11 | eng-fin | ntrex128 | show (logfile) | 20.3 |
| 12 | eng-fin | tatoeba-test-v2020-07-28 | show | 39.4 |
| 13 | eng-fin | tatoeba-test-v2021-03-30 | show | 39.3 |
| 14 | eng-fin | tatoeba-test-v2021-08-07 | show | 39.3 |
| average | 29.647 |