ywmario

Judging from the qps and cost columns, milvus/zilliz is much more efficient! It's probably a good choice for large scale data.

It would be nice to have another column to compare the largest scales these DBs can support.

probably required by some laws

I have been under the impression that the embedded vector is the one actually matters. Token is just another format.

Yes, there will be a lot of 10x engineer

Milvus standalone is pretty neat. I made it successfully run on the first try.

I'm very sure milvus supports it. Milvus calls it hybrid search though: https://milvus.io/docs/hybridsearch.md#Conduct-a-hybrid-vect...