Does Gemini embedding model support languages other than English?

Question

I am planning to use the Gemini embedding model (models/embedding-001) for document/query retrieval, but I can't find anywhere in the documentation if it support languages other than English. Specifically Greek. I wonder if I can get an accurate embedding if my document and query are both in Greek and if my document is in Greek and my query is in English.

Answer 1

If you want to get embeddings for Greek, you can use vertexai's text embedding model textembedding-gecko@001 which supports it. Code example:

from vertexai.language_models import TextEmbeddingModel

def text_embedding() -> list:
    """Text embedding with a Large Language Model."""
    model = TextEmbeddingModel.from_pretrained("textembedding-gecko-multilingual@001")
    embeddings = model.get_embeddings(["What is life?"])
    for embedding in embeddings:
        vector = embedding.values
        print(f"Length of Embedding Vector: {len(vector)}")
    return vector