ID | Benchmark (bleu) | Output | OPUS-MT | bleu | external | bleu | Diff |
---|---|---|---|---|---|---|---|
0 | flores101-devtest | compare | eng-mul/opus2m-2020-08-01 | 9.6 | facebook/nll...illed-1.3B | 27.3 | -17.7 |
1 | flores200-devtest | compare | deu+eng+fra+...2024-05-30 | 27.7 | facebook/nll...illed-1.3B | 27.3 | 0.4 |
2 | tatoeba-test-v2020-07-28 | compare | deu+eng+fra+...2024-05-30 | 8.6 | facebook/nll...illed-1.3B | 13.6 | -5.0 |
3 | tatoeba-test-v2021-03-30 | compare | deu+eng+fra+...2024-05-30 | 8.6 | facebook/nll...illed-1.3B | 13.5 | -4.9 |
4 | tatoeba-test-v2021-08-07 | compare | deu+eng+fra+...2024-05-30 | 8.6 | facebook/nll...illed-1.3B | 13.5 | -4.9 |
average | 12.6 | 19.0 | -6.4 |