Ortix

Latvian-first language infrastructure

Ortix is a growing family of models for Latvian text-processing pipelines: sentence segmentation, correction, punctuation, and normalization. Models built for practical document cleanup, retrieval, and language-system workflows.

Sentence Segmentationortix-seg-lv

Ortix Seg LV

Sentence boundary detection for Latvian documents across academic, legal, parliamentary, web, and scraped text. Built for cleanup, retrieval chunks, correction pipelines, and other downstream usage where punctuation or layout may be unreliable.

Grammatical Error Correctionortix-gec-lv

Ortix GEC LV

Coming soon...

Ortix GEC LV