Back to the catalogue
EmbeddingMultilingualPopular

bge-m3

Multilingual embedding model covering Vietnamese and English in one vector space — no separate index per language.

Ask a question in Vietnamese and bge-m3 will find the English document that answers it, because both were embedded into the same space. A single-language embedding model forces you to keep two indexes and guess which one to search.

This is the default embedding model for a new knowledge base, and it is on every plan — a free workspace has to be able to upload a real document, or there is nothing to evaluate.

Use bge-m3 in your own workspace.

Create a workspace, connect a source, and put this to work in a few minutes.

Book a demo