Model Capabilities
The translation workflow uses only models actually available in the current service; no unconnected brands or fictional scores are shown.
LLM
Used for text translation, terminology processing, and contextual understanding. Actual performance depends on the enabled service.
Vision Models
Used for scanned documents, image text, and visual content recognition. Only works when the relevant service is available.
Specific model names are subject to the options shown in the translation interface at the time of use.