Gives search and RAG backends a serverless vector and BM25 full-text search database built on object storage, queried over an HTTP API or official SDKs.
Use it now
Keep it with up to 9 other tools you're setting up, then share them all as one link.
Try this first
Ask your coding agent:
- Vector Search
Use turbopuffer to run approximate nearest neighbor queries: ‹what you need›
- BM25 Full-Text Search
Use turbopuffer to rank documents: ‹what you need›
- Hybrid Search
Use turbopuffer to combine vector and BM25 results: ‹what you need›
Call the API
Auth An API key from turbopuffer
Its page: turbopuffer.com/docs
Compare
Press Compare to add it to the tray at the bottom of the window, then pick up to 2 others anywhere on the site.
- Access
- API
- Auth
- API Key
- Skills
- 4
Skills
Vector Search
Runs approximate nearest neighbor queries over dense or sparse vectors, combined with attribute filters on an inverted index.
BM25 Full-Text Search
Ranks documents with BM25 full-text search, with ranking and boosting, and supports fast trigram regex queries.
Hybrid Search
Combines vector and BM25 results with server-side reciprocal rank fusion to narrow large candidate sets before re-ranking.
Namespace Branching
Creates instant copy-on-write clones of a namespace with branch_from, with fully independent reads and writes afterward.