turbopuffer

Founded by Simon Hørup Eskildsen in 2023

Run fast vector and full-text search queries directly on cloud object storage

Pricing model

Paid

Rating

★ People love it

Time to value

Quick Setup (< 1 hour)

You can use turbopuffer to run low-latency vector searches and full-text keyword queries directly on top of cloud object storage like Amazon S3. It separates compute from storage, caching active indexes in memory and SSDs while keeping long-term data in inexpensive object storage. This architecture lets you build semantic search, document retrieval, and recommendation systems for generative AI applications at massive scale. You can index billions of documents, filter metadata, execute hybrid BM25 and vector queries, and avoid high infrastructure costs.

What turbopuffer does

Index embedding vectors directly from application workflowsQuery vector namespaces using approximate nearest neighbor algorithmsExecute full-text BM25 queries across document collectionsCombine vector and text search scores in hybrid queriesFilter search queries by document metadata attributesUpsert batches of documents with embedded attributes via APIManage namespace partitions for multi-tenant data storageVector search built directly on cloud object storageIntegrated BM25 full-text keyword searchHybrid search combining dense vectors with sparse keywordsMetadata filtering for namespaces and document attributesIn-memory and SSD multi-tier caching architectureHorizontal scaling supporting over one trillion documentsSub-15ms p50 query latency for warm namespaces

Pricing breakdown

Pricing model: Paid

PlanList price

launch$16 / mo

scale$256 / mo

Tutorials & Demos

Frequently asked

Slack

Want a tailored answer?

See whether turbopuffer fits your stack.

Techbible weighs turbopuffer against what you already pay for, your team shape, and the work that's actually happening. Free to start.

Side by side

Compare turbopuffer

Search the catalog or pick a similar tool to compare pricing, features, and fit.

Or pick from similar tools

More in Data Platform & Integration

All tools →

Ellie.AI

Design conceptual data models and maintain business glossaries collaboratively in your browser

MotherDuck

Run fast SQL analytics in the cloud using DuckDB without managing infrastructure

Acceldata

Monitor data quality, track pipeline performance, and manage enterprise workloads across hybrid environments

AdOpt

Manage website cookie banners and user consent to comply with privacy regulations

Adverity

Extract, harmonize, and deliver marketing data across hundreds of sources to your data warehouse

AGYL AI

Reconcile conflicting business records across different apps into verified data for artificial intelligence

Airbyte

Replicate and sync data across hundreds of sources to warehouses and databases

Aker Systems

Build and run secure cloud data infrastructure for regulated organizations