Logo

Limits

There isn't a limit or performance metric we don't think we can improve by an order of magnitude when prioritized! If you expect to brush up against a limit or are limited by performance by an operation, contact us. Often can be fixed in days.

turbopuffer excels atturbopuffer may not currently be the best fit for
Large scale (1B+ documents/vectors) with lots of namespaces (tens of millions)Large namespaces (100M+)
Naturally sharded data (e.g. B2B where each tenant's data is isolated in its own namespace)Low scale, free tier
Cost-effectivenessAggregation (e.g. group by, sums, explore clusters, ...)
Fast cold startsSingle-digit millisecond latency (tpuf is currently low double digits)
ReliabilityExtensive 1st-stage ranking (we encourage generating a candidate set with hybrid search and refining/re-ranking further in your own 2nd stage)
Hybrid search (BM25 + vector search)Built-in re-ranking (we encourage you to do it in your own application)
Support from DB EngineersBuilt-in embedding (ditto)
Deploy into your VPC (BYOC)Open Source
MetricMax seen in productionProduction limits (current)Production limits (soon)
Max documents (global)35B+Unlimited
Max documents (per namespace)180M100M1B+
Number of namespaces10M+Unlimited
Max dimensions10,752
Max inactive time in cache~3 daysContact us for custom
Write rate (global)100,000 doc/sUnlimited
Write rate (per namespace)10,000 doc/sUnlimited
Max write batch rate (per namespace)1 batch/s1 batch/s4 batches/s
Queries (global)1000+ queries/sUnlimited
Max queries/second (per namespace)100+ queries/s100+ queries/s10,000 QPS
Vector search recall@10~90-95%~90-95%Configurable
Max attribute value8 MiB8 MiB
Max attribute name length128128
Max attributes per document256256
Max namespace name length128128
© 2024 turbopuffer Inc.
Privacy PolicyTerms of service