| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | spa-eng | flores101-devtest | show | 27.6 |
| 1 | spa-eng | flores200-devtest | show | 27.6 |
| 2 | spa-eng | newssyscomb2009 | show | 30.5 |
| 3 | spa-eng | newstest2008 | show (logfile) | 27.9 |
| 4 | spa-eng | newstest2009 | show | 30.3 |
| 5 | spa-eng | newstest2010 | show | 36 |
| 6 | spa-eng | newstest2011 | show | 34.1 |
| 7 | spa-eng | newstest2012 | show | 37.8 |
| 8 | spa-eng | newstest2013 | show | 35.2 |
| 9 | spa-eng | ntrex128 | show (logfile) | 37 |
| 10 | spa-eng | tatoeba-test-v2020-07-28 | show | 59.5 |
| 11 | spa-eng | tatoeba-test-v2021-03-30 | show | 59.8 |
| 12 | spa-eng | tatoeba-test-v2021-08-07 | show | 60.7 |
| 13 | spa-eng | tico19-test | show | 48.4 |
| average | 39.457 |