Turbopuffer
Cheap serverless vector + full-text search on S3
Overview
Turbopuffer is a serverless vector + full-text search database that stores data on S3 — orders of magnitude cheaper than traditional vector DBs for cold data. Used by Notion, Clay, Cursor and others at significant scale.
Use cases
ASEAN Perspective
Turbopuffer in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
Turbopuffer is a serverless vector and full-text search database built on object storage (S3/GCS/Azure) with a hot NVMe/memory cache, delivering hybrid BM25 + vector search at low cost and sub-10ms warm p50 latency. Its object-storage-first design makes it unusually cheap at scale with no hard namespace limits, a strong fit for multi-tenant AI/RAG workloads.
It suits engineering teams building search/RAG who want predictable, low storage cost over keeping everything in RAM. Caveats: cold-query latency on infrequently accessed data is the architectural trade-off (critics note tail-latency cost), and it is a younger product than Pinecone/pgvector with a smaller ecosystem. It offers a clean HTTP API plus TypeScript, Python and Go SDKs. Usage-based pricing from ~$64/mo; cloud-region availability should be checked for ASEAN latency/residency.
About this listing
This entry was compiled from publicly available data including Turbopuffer's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Turbopuffer unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Turbopuffer directly →
Spotted something out of date? Suggest an update →
Alternatives to Turbopuffer
More in Code & Dev Tools