docs/integrations/vectorstores/

[giscus[bot]]

docs/integrations/vectorstores/

A vector store stores embedded data and performs similarity search.

https://python.langchain.com/docs/integrations/vectorstores/

[tjmcdonough]

I think QdrantVectorStore does support multitenancy?

[dov84d]

What do you mean "Search by Vector"?
How is that, that some Vector stores doesn't have the capability to search by vector?

[hesreallyhim]

I think it means "search by similarity to a passed-in vector/embedding" as opposed to "search by similarity to a document"

[tang-john]

Chroma supports Doc IDs during add. It's even documented on this site https://python.langchain.com/docs/integrations/vectorstores/chroma/#add-items-to-vector-store

[hesreallyhim]

Yeah I wasn't clear what "IDs in add Documents" means... I inferred likewise that it means that Documents with IDs will have those IDs added as the IDs in the vectorstore... if so, the API reference for langchain-chroma at least heavily implies that it supports this functionality:

Parameters
:
documents (list[Document]) – Documents to add to the vectorstore.

kwargs (Any) – Additional keyword arguments. if kwargs contains ids and documents contain ids, the ids in the kwargs will receive precedence.