turbopuffer
A serverless vector and full-text search engine built on object storage for low cost.
turbopuffer is built on a different cost model: it keeps data on cheap object storage and caches hot data, rather than holding everything on expensive memory or SSD. The result is dramatically lower cost for large but not constantly-hot datasets.
It does vector and full-text search together, and is fully managed.
Where it fits
A fit for large search workloads where cost matters and not all of the data is hot at once.
Where it does not fit
A managed service: not the choice when data must stay self-hosted; and pure in-memory stores still win on raw latency.
Alternatives to turbopuffer
The closest options on the map, and where each fits instead.
-
Pinecone Managed vector database
A fit when a team wants vector search with zero operational burden and a hosted service is acceptable.
Unlike turbopuffer Maturity: established
-
Qdrant Dedicated vector database
A strong pick when retrieval needs to combine similarity with metadata filtering, or has outgrown an in-database extension.
Unlike turbopuffer Open sourceSelf-hostableMaturity: established
Pairs with
Commonly used alongside it.
