Find answers from the community

Updated 11 months ago

Hi, I’m trying to save documents that

Hi, I’m trying to save documents that have been embedded to mongodb, but I don’t know how to persist using an ingestion pipeline
R
T
L
8 comments
If you use remote services like mongo or redis, then persisting is not needed
@Rohan but what about embeddings, do I need to redo them each time I use it ?
If the embeddings are stored in mongodb, like you mentioned, then they'll be fetched from there during querying
@Rohan but how can you store the embeddings in mongodb ?
Mongodb might be the worst vector db (since it requires a manual step to create the "index" in the GUI before you can query it)

But there is a small example here
https://docs.llamaindex.ai/en/stable/examples/vector_stores/MongoDBAtlasVectorSearch.html
I would use another vector db tbh
Thanks @Logan M would elastic search be better ? Or what are the OSS I can self host ?
qdrant, chroma, elastic, redis, postgres are all good self-host options
Add a reply
Sign up and join the conversation on Discord