Is this multilingual and has this been tested on non-English only benchmarks?
#16
by
Zoahmed1
- opened
Wondering if this model can be used for embedding non-English documents.
Have you any news about this?
This model "NV-Embed-v2" is only for English only!
No multi-lingual support. For that you either can train your own model using the methods specified in the paper linked here: https://arxiv.org/pdf/2405.17428
Or consider some alternative multi-lingual embedding models like jina-embeddings-v3.
I hope this answers your question.