| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | eng-spa | flores101-devtest | show | 28.5 |
| 1 | eng-spa | flores200-devtest | show | 28.5 |
| 2 | eng-spa | newssyscomb2009 | show | 31.5 |
| 3 | eng-spa | newstest2008 | show (logfile) | 30.1 |
| 4 | eng-spa | newstest2009 | show | 30.2 |
| 5 | eng-spa | newstest2010 | show | 37.6 |
| 6 | eng-spa | newstest2011 | show | 38.9 |
| 7 | eng-spa | newstest2012 | show | 39.5 |
| 8 | eng-spa | newstest2013 | show | 35.9 |
| 9 | eng-spa | ntrex128 | show (logfile) | 42.4 |
| 10 | eng-spa | tatoeba-test-v2020-07-28 | show | 55 |
| 11 | eng-spa | tatoeba-test-v2021-03-30 | show | 55.8 |
| 12 | eng-spa | tatoeba-test-v2021-08-07 | show | 57.2 |
| 13 | eng-spa | tico19-test | show | 53 |
| average | 40.293 |