List your company

Intrinsic tokenization quality evaluation on CulturaX 18 languages sample

4.89WFText
3.11163.57334.0354.4967Jun 18, 2026

Evaluation results

MethodLinks
TextVocabulary size=300k, Aggregation=macro-averaged2026.064.89922.02138.816.373.992.2620.293.336
IPAVocabulary size=300k, Aggregation=macro-averaged2026.063.18902.72169.146.063.72.9820.161.7217