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 |
|---|---|---|---|---|---|---|
| French → PortugueseFR · French · Français→PT · Portuguese · Português | 1 | Median 9.03 Mean 7.80 · 25th pct. 8.79 · 75th pct. 9.23 | Mean 0.844 25th pct. 0.842 · Median 0.846 · 75th pct. 0.847 | Mean 27.8 25th pct. 25.7 · Median 27.1 · 75th pct. 29.3 | Mean 32.9 25th pct. 32.3 · Median 32.8 · 75th pct. 33.7 | Mean 0.940 25th pct. 0.930 · Median 0.937 · 75th pct. 0.952 |
| French → EnglishFR · French · Français→EN · English | 2 | Median 8.95 Mean 7.79 · 25th pct. 8.53 · 75th pct. 9.10 | Mean 0.846 25th pct. 0.844 · Median 0.846 · 75th pct. 0.847 | Mean 22.0 25th pct. 19.9 · Median 22.0 · 75th pct. 24.4 | Mean 25.2 25th pct. 23.2 · Median 25.8 · 75th pct. 27.2 | Mean 1.086 25th pct. 1.083 · Median 1.087 · 75th pct. 1.092 |
| Russian → PortugueseRU · Russian · Русский→PT · Portuguese · Português | 3 | Median 8.91 Mean 7.89 · 25th pct. 8.31 · 75th pct. 9.22 | Mean 0.822 25th pct. 0.815 · Median 0.820 · 75th pct. 0.833 | Mean 22.3 25th pct. 17.4 · Median 20.7 · 75th pct. 26.6 | Mean 31.0 25th pct. 29.0 · Median 30.9 · 75th pct. 33.4 | Mean 1.082 25th pct. 1.074 · Median 1.081 · 75th pct. 1.090 |
| Hindi → EnglishHI · Hindi · हिन्दी→EN · English | 4 | Median 8.89 Mean 7.83 · 25th pct. 8.51 · 75th pct. 9.19 | Mean 0.849 25th pct. 0.845 · Median 0.850 · 75th pct. 0.855 | Mean 21.1 25th pct. 18.8 · Median 19.9 · 75th pct. 21.2 | Mean 27.3 25th pct. 26.5 · Median 27.1 · 75th pct. 28.0 | Mean 0.972 25th pct. 0.957 · Median 0.971 · 75th pct. 0.980 |
| Hindi → FrenchHI · Hindi · हिन्दी→FR · French · Français | 5 | Median 8.88 Mean 7.93 · 25th pct. 8.42 · 75th pct. 9.11 | Mean 0.858 25th pct. 0.856 · Median 0.859 · 75th pct. 0.863 | Mean 27.1 25th pct. 22.8 · Median 24.7 · 75th pct. 26.5 | Mean 38.7 25th pct. 37.6 · Median 38.2 · 75th pct. 39.7 | Mean 0.870 25th pct. 0.861 · Median 0.870 · 75th pct. 0.875 |
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.