Loading benchmarks…
Benchmark dashboard
Automated translation metrics summarized for 335 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 |
|---|---|---|---|---|---|---|
| English → ThaiEN · English→TH · Thai · ไทย | 298 | Median 5.00 Mean 4.98 · 25th pct. 5.00 · 75th pct. 5.00 | Mean 0.939 25th pct. 0.922 · Median 0.931 · 75th pct. 0.957 | Mean 0.0 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.0 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.199 25th pct. 0.170 · Median 0.170 · 75th pct. 0.225 |
| English → KazakhEN · English→KK · Kazakh · Қазақша | 170 | Median 7.14 Mean 6.27 · 25th pct. 6.54 · 75th pct. 7.74 | Mean 0.914 25th pct. 0.922 · Median 0.929 · 75th pct. 0.932 | Mean 10.4 25th pct. 9.0 · Median 9.9 · 75th pct. 10.5 | Mean 15.0 25th pct. 14.1 · Median 15.2 · 75th pct. 16.2 | Mean 0.845 25th pct. 0.822 · Median 0.833 · 75th pct. 0.862 |
| English → RomanianEN · English→RO · Romanian · Română | 66 | Median 8.15 Mean 7.11 · 25th pct. 7.62 · 75th pct. 8.29 | Mean 0.925 25th pct. 0.921 · Median 0.928 · 75th pct. 0.930 | Mean 38.1 25th pct. 34.8 · Median 36.2 · 75th pct. 39.9 | Mean 45.0 25th pct. 43.4 · Median 44.6 · 75th pct. 47.3 | Mean 1.005 25th pct. 0.998 · Median 1.001 · 75th pct. 1.016 |
| Mandarin Chinese → JapaneseZH · Mandarin Chinese · 普通话→JA · Japanese · 日本語 | 314 | Median 4.27 Mean 3.85 · 25th pct. 3.38 · 75th pct. 4.77 | Mean 0.921 25th pct. 0.919 · Median 0.923 · 75th pct. 0.924 | Mean 0.0 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.0 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.842 25th pct. 0.778 · Median 0.778 · 75th pct. 0.778 |
| Portuguese → RomanianPT · Portuguese · Português→RO · Romanian · Română | 70 | Median 8.13 Mean 7.36 · 25th pct. 7.43 · 75th pct. 8.85 | Mean 0.916 25th pct. 0.911 · Median 0.922 · 75th pct. 0.923 | Mean 33.9 25th pct. 28.9 · Median 33.9 · 75th pct. 37.7 | Mean 43.3 25th pct. 41.8 · Median 43.9 · 75th pct. 46.2 | Mean 1.150 25th pct. 1.142 · 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.