| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | eng-spa | flores101-devtest | show | 24.4 |
| 1 | eng-spa | newssyscomb2009 | show | 29.9 |
| 2 | eng-spa | newstest2009 | show | 29 |
| 3 | eng-spa | newstest2010 | show | 34.4 |
| 4 | eng-spa | newstest2011 | show | 36.1 |
| 5 | eng-spa | newstest2012 | show | 36.1 |
| 6 | eng-spa | newstest2013 | show | 33.1 |
| 7 | eng-spa | tatoeba-test-v2020-07-28 | show | 52 |
| 8 | eng-spa | tatoeba-test-v2021-03-30 | show | 52.9 |
| 9 | eng-spa | tatoeba-test-v2021-08-07 | show | 53.8 |
| 10 | eng-spa | tico19-test | show | 46.5 |
| average | 38.927 |