Muvera: Making multi-vector retrieval as fast as single-vector search

51 georgehill 2 6/26/2025, 10:29:34 AM research.google ↗

Comments (2)

dinobones · 5m ago
So this is basically an “embedding of embeddings”, an approximation of multiple embeddings compressed into one, to reduce dimensionality/increase performance.

All this tells me is that: the “multiple embeddings” are probably mostly overlapping and the marginal value of each additional one is probably low, if you can represent them with a single embedding.

I don’t otherwise see how you can keep comparable performance without breaking information theory.

trengrj · 2h ago
We added Muvera to Weaviate recently https://weaviate.io/blog/muvera and also have a nice podcast on it https://www.youtube.com/watch?v=nSW5g1H4zoU.

When looking at multi-vector / ColBERT style approaches, the embedding per token approach can massively increase costs. You might go from a single 768 dimension vector to 128 x 130 = 16,640 dimensions. Even with better results from a multi-vector model this can make it unfeasible for many use-cases.

Muvera, converts the multiple vectors into a single fixed dimension (usually net smaller) vector that can be used by any ANN index. As you now have a single vector you can use all your existing ANN algorithms and stack other quantization techniques for memory savings. In my opinion it is a much better approach than PLAID because it doesn't require specific index structures or clustering assumptions and can achieve lower latency.