The ModelFront machine translation quality prediction API provides a segment-level quality score, so you can focus human translators only where they’re needed most. The ModelFront API predicts if a machine translation is good or bad. It learns from your post-editing data, to reflect your domain, terminology, style – even for specific brands, products or quality tiers. ModelFront works with any translation management system and any machine translation, from Microsoft Custom Translator to DeepL to large language models (LLMs) like ChatGPT.