State-of-the-art large embedding model from Mixedbread AI, ranked first among similar-sized models, supporting Matryoshka Representation Learning and binary quantization with 700M+ training pairs.
mxbai-embed-large is a state-of-the-art large embedding model from mixedbread.ai. It's part of the crispy sentence embedding family from Mixedbread.
The model:
The model was trained with:
The model supports Matryoshka Representation Learning, allowing vector truncation to smaller dimensions without retraining.
Supports binary quantization for reduced storage and faster similarity search.
For retrieval tasks, you need to provide the prompt "Represent this sentence for searching relevant passages:" for query.
Free and open-source.
Loading more......