Loading benchmarks…
Automated translation metrics summarized for 334 language directions.
Sources — Source segments come from curated, public-facing parallel corpora and benchmark-style collections mixed for broad coverage. They are not model-generated: originals and references are assembled upstream as parallel text only; this dashboard does not redistribute raw corpora.
Mixed parallel corpora — Parallel segments spanning diverse genres and corpus scales. Each item includes a reference translation that is certified and human-verified—not produced by the models being benchmarked.
Language pairi | Rankingi | Fluencyi | COMETi | BLEUi | sacreBLEUi | Len ratioi |
|---|---|---|---|---|---|---|
| Spanish → EnglishES · Spanish · Español→EN · English | 171 | Mean 7.42 25th pct. 7.20 · Median 7.51 · 75th pct. 7.68 | Mean 0.881 25th pct. 0.880 · Median 0.882 · 75th pct. 0.884 | Mean 49.8 25th pct. 48.4 · Median 50.2 · 75th pct. 52.1 | Mean 53.4 25th pct. 51.9 · Median 54.3 · 75th pct. 55.1 | Mean 1.158 25th pct. 1.149 · Median 1.169 · 75th pct. 1.173 |
| English → VietnameseEN · English→VI · Vietnamese · Tiếng Việt | 79 | Mean 2.19 25th pct. 2.03 · Median 2.23 · 75th pct. 2.40 | Mean 0.911 25th pct. 0.908 · Median 0.912 · 75th pct. 0.915 | Mean 48.1 25th pct. 46.2 · Median 47.5 · 75th pct. 49.3 | Mean 52.2 25th pct. 50.3 · Median 51.4 · 75th pct. 53.6 | Mean 1.567 25th pct. 1.526 · Median 1.577 · 75th pct. 1.603 |
| Portuguese → EnglishPT · Portuguese · Português→EN · English | 51 | Mean 6.92 25th pct. 6.43 · Median 6.88 · 75th pct. 7.29 | Mean 0.896 25th pct. 0.895 · Median 0.896 · 75th pct. 0.898 | Mean 40.7 25th pct. 39.8 · Median 40.9 · 75th pct. 43.0 | Mean 45.7 25th pct. 44.9 · Median 46.2 · 75th pct. 48.2 | Mean 1.213 25th pct. 1.206 · Median 1.212 · 75th pct. 1.222 |
| English → PortugueseEN · English→PT · Portuguese · Português | 125 | Mean 7.06 25th pct. 6.89 · Median 7.07 · 75th pct. 7.29 | Mean 0.867 25th pct. 0.859 · Median 0.865 · 75th pct. 0.872 | Mean 31.8 25th pct. 29.1 · Median 32.0 · 75th pct. 34.2 | Mean 44.2 25th pct. 42.0 · Median 44.6 · 75th pct. 47.2 | Mean 0.901 25th pct. 0.894 · Median 0.896 · 75th pct. 0.907 |
| Portuguese → RomanianPT · Portuguese · Português→RO · Romanian · Română | 79 | Mean 7.24 25th pct. 6.58 · Median 7.47 · 75th pct. 7.80 | Mean 0.918 25th pct. 0.911 · Median 0.922 · 75th pct. 0.923 | Mean 32.3 25th pct. 28.9 · Median 33.3 · 75th pct. 35.6 | Mean 44.1 25th pct. 41.8 · Median 44.5 · 75th pct. 46.2 | Mean 1.149 25th pct. 1.136 · Median 1.153 · 75th pct. 1.161 |
This release reports translation quality by language pair: medians and spread of automated scores (fluency, COMET, BLEU, sacreBLEU, length ratio) aggregated across evaluated directions. Run metadata describes recipes, metrics, and segment counts.