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 |
|---|---|---|---|---|---|---|
| Modern Standard Arabic → JapaneseAR · Modern Standard Arabic · العربية الفصحى→JA · Japanese · 日本語 | 74 | Median 7.03 Mean 7.05 · 25th pct. 6.71 · 75th pct. 7.43 | Mean 0.827 25th pct. 0.821 · Median 0.830 · 75th pct. 0.835 | Mean 0.1 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 1.0 25th pct. 0.0 · Median 0.0 · 75th pct. 0.7 | Mean 0.124 25th pct. 0.118 · Median 0.118 · 75th pct. 0.131 |
| Modern Standard Arabic → Yue Chinese (Cantonese)AR · Modern Standard Arabic · العربية الفصحى→YUE · Yue Chinese (Cantonese) · 粵語 | 17 | Median 7.64 Mean 7.36 · 25th pct. 7.19 · 75th pct. 7.76 | Mean 0.299 25th pct. 0.295 · Median 0.301 · 75th pct. 0.303 | Mean 0.1 25th pct. 0.0 · Median 0.0 · 75th pct. 0.1 | Mean 0.4 25th pct. 0.0 · Median 0.0 · 75th pct. 0.3 | Mean 0.188 25th pct. 0.170 · Median 0.178 · 75th pct. 0.210 |
| English → ThaiEN · English→TH · Thai · ไทย | 300 | Median 2.86 Mean 2.84 · 25th pct. 2.86 · 75th pct. 2.86 | Mean 0.937 25th pct. 0.922 · Median 0.925 · 75th pct. 0.956 | 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 0.198 25th pct. 0.170 · Median 0.170 · 75th pct. 0.198 |
| English → Mandarin ChineseEN · English→ZH · Mandarin Chinese · 普通话 | 306 | Median 2.71 Mean 2.65 · 25th pct. 2.47 · 75th pct. 2.86 | Mean 0.861 25th pct. 0.860 · Median 0.865 · 75th pct. 0.869 | Mean 0.2 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.9 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.029 25th pct. 0.025 · Median 0.025 · 75th pct. 0.025 |
| Spanish → JapaneseES · Spanish · Español→JA · Japanese · 日本語 | 39 | Median 7.42 Mean 7.34 · 25th pct. 6.93 · 75th pct. 7.81 | Mean 0.813 25th pct. 0.802 · Median 0.820 · 75th pct. 0.830 | Mean 0.0 25th pct. 0.0 · Median 0.0 · 75th pct. 0.0 | Mean 0.4 25th pct. 0.0 · Median 0.0 · 75th pct. 0.1 | Mean 0.136 25th pct. 0.100 · Median 0.115 · 75th pct. 0.144 |
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.