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 |
|---|---|---|---|---|---|---|
| Mandarin Chinese → VietnameseZH · Mandarin Chinese · 普通话→VI · Vietnamese · Tiếng Việt | 125 | Mean 2.41 25th pct. 2.26 · Median 2.34 · 75th pct. 2.64 | Mean 0.875 25th pct. 0.869 · Median 0.876 · 75th pct. 0.881 | Mean 37.5 25th pct. 35.6 · Median 37.5 · 75th pct. 39.5 | Mean 39.5 25th pct. 37.7 · Median 39.8 · 75th pct. 41.4 | Mean 50.561 25th pct. 50.194 · Median 50.833 · 75th pct. 51.000 |
| Mandarin Chinese → Modern Standard ArabicZH · Mandarin Chinese · 普通话→AR · Modern Standard Arabic · العربية الفصحى | 9 | Mean 2.76 25th pct. 2.70 · Median 2.84 · 75th pct. 2.86 | Mean 0.802 25th pct. 0.800 · Median 0.803 · 75th pct. 0.805 | Mean 20.4 25th pct. 16.9 · Median 20.6 · 75th pct. 24.7 | Mean 22.0 25th pct. 16.8 · Median 22.8 · 75th pct. 27.3 | Mean 41.167 25th pct. 40.500 · Median 41.000 · 75th pct. 42.167 |
| Mandarin Chinese → Iranian Persian (Farsi)ZH · Mandarin Chinese · 普通话→FA · Iranian Persian (Farsi) · فارسی | 79 | Mean 2.54 25th pct. 2.41 · Median 2.52 · 75th pct. 2.71 | Mean 0.863 25th pct. 0.857 · Median 0.864 · 75th pct. 0.868 | Mean 16.2 25th pct. 14.6 · Median 16.3 · 75th pct. 17.6 | Mean 17.0 25th pct. 15.7 · Median 16.9 · 75th pct. 17.7 | Mean 35.106 25th pct. 33.833 · Median 35.000 · 75th pct. 35.944 |
| Mandarin Chinese → IndonesianZH · Mandarin Chinese · 普通话→ID · Indonesian · Bahasa Indonesia | 79 | Mean 2.57 25th pct. 2.43 · Median 2.62 · 75th pct. 2.71 | Mean 0.880 25th pct. 0.877 · Median 0.881 · 75th pct. 0.882 | Mean 13.0 25th pct. 11.1 · Median 13.1 · 75th pct. 14.7 | Mean 18.1 25th pct. 16.0 · Median 17.8 · 75th pct. 19.9 | Mean 32.728 25th pct. 31.972 · Median 32.667 · 75th pct. 33.139 |
| Mandarin Chinese → TurkishZH · Mandarin Chinese · 普通话→TR · Turkish · Türkçe | 125 | Mean 2.49 25th pct. 2.38 · Median 2.51 · 75th pct. 2.62 | Mean 0.850 25th pct. 0.846 · Median 0.852 · 75th pct. 0.854 | Mean 8.5 25th pct. 6.9 · Median 8.5 · 75th pct. 9.5 | Mean 10.6 25th pct. 8.8 · Median 11.3 · 75th pct. 12.1 | Mean 27.689 25th pct. 27.306 · Median 27.778 · 75th pct. 28.333 |
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.