← Data and charts
AI & evaluation · Updated 2026-11-04

Hosted models score highest on Nigerian-language translation; N-ATLaS is well ahead of the open baseline

Mean chrF++ across six translation directions (English with Hausa, Igbo and Yoruba). Higher is better.

Hover, focus or click a bar for its detail. Click to pin it and compare the others against it.

Unit: chrF++. 100 aligned sentence groups; model labels and scores are as recorded by the project, with no independent rerun and no confidence intervals. chrF++ is an overlap score, not percent accuracy.

Cite this chart

Extrafemi Research (2026). Hosted models score highest on Nigerian-language translation; N-ATLaS is well ahead of the open baseline [Chart]. Data from Naija Eval: An Open Evaluation Pipeline for Nigerian-Language Translation and Speech Recognition. https://research.extrafemi.com/data/naija-eval-translation-scores