@jjacky@pinecone Nah this is a diff index type, it doesn't generate or do chunking for you like assistant does! The idea is that you don't need to have several indexes anymore, you can have one Documents API index that supports the different search types. Documents bc the each unit has a schema
Boom! All you need is @pinecone to build all sorts of search now.
One API key, and one Documents API index means you can do dense/semantic, sparse, keyword/fts search on the same index. It's never been easier to add knowledge to your apps.
Try it out!
Your search stack shouldn't need its own embedding pipeline. Send plain text to the Pinecone Documents API, and one field serves both semantic and full-text search. Pinecone embeds your documents and queries with the same model.
https://t.co/9QnVQQnmdC
Hybrid search used to take three indexes. Now it takes one.
New in the Pinecone changelog:
→ Hybrid search in one index
→ BYOC is GA, now with on-demand
→ VQ-Bench for vector quantizers
→ Cost and usage export in FOCUS format
Watch the full changelog 👇
https://t.co/OHxqZA0iNo
made with @HyperFrames_ shoutout @Miguel07Code
I made a tiny live-editor for my website so the copy was easier to change!
Just ask for the tools you need