Expected Qdrant latency

[RemainIndoors1]

I'm pretty new to Qdrant and vector storage in general, but I'm just wondering what the normal response speed is supposed to be when querying for points in a collection from python qdrant client. I have Qdrant installed in Docker Desktop with a single collection that has 6 points stored, and even following the documentation's recommendations for lowering latency, I can't get a response for 5 points in less than 2 seconds. The model generating embeddings takes 0.02 seconds, but querying Qdrant is painfully slow for me at the moment. Is this normal? or am I doing something wrong?

[generall]

no, it is not normal.

[RemainIndoors1]

Alright, thanks for the info.

[RemainIndoors1]

Ahh, it looks like that 2 seconds was the difference between prefer_grpc=False and prefer_grpc=True
I set that to true and the same queries went from 2.04s to 0.01 seconds.

[generall]

for both parameters 2 seconds is too much.