| ID | Language | Benchmark | Output | chrf |
|---|---|---|---|---|
| 0 | deu-eng | flores101-devtest | show | 0.694 |
| 1 | deu-eng | flores200-devtest | show | 0.694 |
| 2 | deu-eng | multi30k_task2_test_2016 | show | 0.21926 |
| 3 | deu-eng | multi30k_test_2016_flickr | show | 0.62811 |
| 4 | deu-eng | multi30k_test_2017_flickr | show | 0.66164 |
| 5 | deu-eng | multi30k_test_2017_mscoco | show | 0.59329 |
| 6 | deu-eng | multi30k_test_2018_flickr | show | 0.62143 |
| 7 | deu-eng | newssyscomb2009 | show | 0.55936 |
| 8 | deu-eng | newstest2009 | show | 0.55242 |
| 9 | deu-eng | newstest2010 | show | 0.60475 |
| 10 | deu-eng | newstest2011 | show | 0.56576 |
| 11 | deu-eng | newstest2012 | show | 0.59023 |
| 12 | deu-eng | newstest2013 | show | 0.60836 |
| 13 | deu-eng | newstest2014 | show | 0.63382 |
| 14 | deu-eng | newstest2015 | show | 0.64003 |
| 15 | deu-eng | newstest2016 | show | 0.69123 |
| 16 | deu-eng | newstest2017 | show | 0.65779 |
| 17 | deu-eng | newstest2018 | show | 0.70248 |
| 18 | deu-eng | newstest2019 | show | 0.65172 |
| 19 | deu-eng | newstest2020 | show | 0.65707 |
| 20 | deu-eng | newstestB2020 | show | 0.65534 |
| 21 | deu-eng | tatoeba-test-v2020-07-28 | show | 0.70302 |
| 22 | deu-eng | tatoeba-test-v2021-03-30 | show | 0.69366 |
| 23 | deu-eng | tatoeba-test-v2021-08-07 | show | 0.68552 |
| average | 0.624 |