Loading benchmarks…
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 | fluency_v2.0i | COMETi | BLEUi | sacreBLEUi | Len ratioi |
|---|---|---|---|---|---|---|
| English → BhojpuriEN · English→BHO · Bhojpuri · भोजपुरी | 1 | Mean 2.12 25th pct. 2.12 · Median 2.12 · 75th pct. 2.12 | Mean 0.623 25th pct. 0.623 · Median 0.623 · 75th pct. 0.623 | Mean 2.0 25th pct. 2.0 · Median 2.0 · 75th pct. 2.0 | — | Mean 1.265 25th pct. 1.265 · Median 1.265 · 75th pct. 1.265 |
| French → VietnameseFR · French · Français→VI · Vietnamese · Tiếng Việt | 9 | Mean 3.75 25th pct. 3.56 · Median 3.77 · 75th pct. 3.92 | Mean 0.876 25th pct. 0.866 · Median 0.875 · 75th pct. 0.884 | Mean 22.5 25th pct. 20.4 · Median 22.8 · 75th pct. 23.6 | Mean 30.0 25th pct. 28.5 · Median 29.8 · 75th pct. 31.3 | Mean 1.294 25th pct. 1.276 · Median 1.290 · 75th pct. 1.304 |
| Russian → Iranian Persian (Farsi)RU · Russian · Русский→FA · Iranian Persian (Farsi) · فارسی | 29 | Mean 3.88 25th pct. 3.73 · Median 3.81 · 75th pct. 4.13 | Mean 0.867 25th pct. 0.863 · Median 0.866 · 75th pct. 0.874 | Mean 17.4 25th pct. 16.2 · Median 17.8 · 75th pct. 18.1 | Mean 24.1 25th pct. 23.6 · Median 24.4 · 75th pct. 24.9 | Mean 1.254 25th pct. 1.220 · Median 1.261 · 75th pct. 1.283 |
| Mandarin Chinese → ThaiZH · Mandarin Chinese · 普通话→TH · Thai · ไทย | 266 | Mean 3.78 25th pct. 3.59 · Median 3.81 · 75th pct. 4.04 | Mean 0.557 25th pct. 0.550 · Median 0.561 · 75th pct. 0.570 | 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 2.439 25th pct. 2.222 · Median 2.389 · 75th pct. 2.694 |
| English → Iranian Persian (Farsi)EN · English→FA · Iranian Persian (Farsi) · فارسی | 27 | Mean 4.10 25th pct. 3.79 · Median 4.00 · 75th pct. 4.55 | Mean 0.897 25th pct. 0.897 · Median 0.900 · 75th pct. 0.903 | Mean 24.0 25th pct. 19.9 · Median 22.4 · 75th pct. 27.3 | Mean 25.0 25th pct. 20.9 · Median 23.8 · 75th pct. 28.0 | Mean 1.189 25th pct. 1.141 · Median 1.189 · 75th pct. 1.231 |
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.