#multilingual-models

#multilingual-models

Hugging Face has released mmBERT, a new multilingual encoder trained on more than 3 trillion tokens across 1,833 languages. The model builds on the ModernBERT architecture and is the first to significantly improve upon XLM-R, a long-time baseline for multilingual understanding tasks. mmBERT uses a progressive training schedule instead of training on all languages at once. It starts with 60 high-resource languages, expands to 110, and finally includes all 1,833 languages.

Mistral's Voxtral speech models offer an alternative to closed APIs, combining high accuracy, multilingual support, and extensive context processing at competitive prices.