Skip to main content
Version: Next

Multilingual E5 Large Instruct

Multilingual E5 Large Instruct is an embedding model that was initialized from xlm-roberta-large and continually trained on a mixture of multilingual datasets. It supports 100 languages from xlm-roberta.

Model ID​

intfloat/multilingual-e5-large-instruct

Source​

Hugging face

Modality​

  • Input: text
  • Output: embedding vector

Context limit​

  • Embedding size: 1024; the model doesn't support reduced dimensions
  • Maximum input size: 512 tokens; longer inputs are truncated

Endpoints​