We designed the tpuf v3 query engine to benefit from vectorization everywhere. The first part of the engine to benefit from this is our filter evaluation logic.

Follow along as we evolve turbopuffer's storage architecture and grind latency down til it's better than it ever was

turbopuffer is pushing the frontier of search. To do that, we have to fundamentally redesign our storage architecture so the vector index is no longer primary.

Native embedding isn't just a convenience feature, but a way to shave hundreds of milliseconds off your search queries.

Kepler builds verifiable AI for investment firms and financial analysts. With turbopuffer they can search every SEC filing, earnings transcript, and SharePoint import at far lower cost than alternatives.

turbopuffer has improved text search performance up to 20x by adopting the same text search algorithm as Apache Lucene, a vectorized variant of block-max MAXSCORE

Applied Compute partnered with turbopuffer to show that a small, specialized search model post-trained to use precomputed search indexes can achieve impressive search quality at a fraction of the cost and latency of frontier models.

fal shipped search in Assets so users and their agents can find and reference previously generated media assets. turbopuffer runs filtered hybrid search over 300M+ videos, images, and 3D models at millisecond latency.

We deploy many database upgrades every day across ~100 public, single-tenant, and BYOC clusters. It would be insane to use Terraform or Helm to do that, so we built our own control plane.

Postgres didn't scale for Mem0's agent memory platform. They migrated hundreds of millions of memories to turbopuffer, reducing end-to-end latency by 70x.

Legora replaced Elasticsearch and pgvector + DiskANN with turbopuffer to build an efficient retrieval layer that minimizes input tokens without sacrificing performance, all with legal-grade data isolation and security.

Atlassian chose turbopuffer for its operational and architectural simplicity, cloud-agnostic BYOC deployment model, low cost, and virtually unlimited scalability. It now underpins search for millions of Atlassian customers on Jira, Confl...

SID-1 is an agentic search model that is 24x faster than GPT-5.1-high, 374x cheaper than Sonnet 4.5, and achieves 1.9x higher recall than traditional RAG pipelines. Here's how we trained it using large-scale RL on turbopuffer.