| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | rus-eng | flores101-devtest | show | 26.7 |
| 1 | rus-eng | newstest2012 | show | 33.1 |
| 2 | rus-eng | newstest2013 | show | 26.2 |
| 3 | rus-eng | newstest2014 | show | 29.5 |
| 4 | rus-eng | newstest2015 | show | 28.5 |
| 5 | rus-eng | newstest2016 | show | 27.9 |
| 6 | rus-eng | newstest2017 | show | 31.1 |
| 7 | rus-eng | newstest2018 | show | 27.2 |
| 8 | rus-eng | newstest2019 | show | 29.2 |
| 9 | rus-eng | newstest2020 | show | 26.9 |
| 10 | rus-eng | newstestB2020 | show | 28.5 |
| 11 | rus-eng | tatoeba-test-v2020-07-28 | show | 54.6 |
| 12 | rus-eng | tatoeba-test-v2021-03-30 | show | 54.2 |
| 13 | rus-eng | tatoeba-test-v2021-08-07 | show | 52.6 |
| 14 | rus-eng | tico19-test | show | 25.8 |
| average | 33.467 |