| ID | Language | Benchmark | Output | bleu |
|---|---|---|---|---|
| 0 | yue-eng | flores200-devtest | show | 21.1 |
| 1 | yue-eng | ntrex128 | show (logfile) | 18 |
| 2 | zho-eng | generaltest2022 | show (logfile) | 19.2 |
| 3 | zho-eng | newstest2017 | show | 19.8 |
| 4 | zho-eng | newstest2018 | show | 19.9 |
| 5 | zho-eng | newstest2019 | show | 21.8 |
| 6 | zho-eng | newstest2020 | show | 21.9 |
| 7 | zho-eng | newstest2021 | show | 19.2 |
| 8 | zho-eng | newstestALL2020 | show (logfile) | 28.9 |
| 9 | zho-eng | newstestB2020 | show | 17.8 |
| 10 | zho-eng | tatoeba-test-v2020-07-28 | show | 35.7 |
| 11 | zho-eng | tatoeba-test-v2021-03-30 | show | 35.6 |
| 12 | zho-eng | tatoeba-test-v2021-08-07 | show | 35.6 |
| 13 | zho-eng | tico19-test | show | 25.7 |
| average | 24.300 |