Pinecone, the leading closed-source vector database provider, is known for being fast, scalable, and easy to use. Its ability to allow users to per...
For further actions, you may consider blocking this person and/or reporting abuse
Nice article. Very informative. Can I ask why did you pass up other solutions such as weaviate, jina and chromadb? Was it for the sake of simplicity since pgvector allows you to rely on just one database solution? I'm curious to know how or if you evaluated those solutions and take us through you/ your team's reasoning. Many thanks.
Definitely for the sake of simplicity, for Weaviate I'll have to manage my own infrastructure and if I were to use WCS then I run into a fragmented data source problem (although Weaviate doesn't have a metadata limit). I didn't consider chroma because they were still in-memory only at the time, and Jina frankly I didn't know about them.
At the end of the day, there was just no need to have another standalone vector database, otherwise we would have assessed all solutions in much more detail. Hope this helps!
It certainly does help to hear these thoughts. I am actually in the middle of evaluating the entire list of vector dB solutions for my own project so this very helpful. Many thanks 🙏
Good read. Thank you. I work on the open source vector database, Milvus. I am always looking for feedback to make our project better - even if it is feedback like you don't want another db to manage.
PGVector is really cool!
It sure is :)
I created a simple wrapper to pgvector that is inspired by Pinecone's simplicity: github.com/UpMortem/simple-pgvecto...