Warning: Queries may be slow during periods of high write throughput or after a large bulk import.

turbopuffer can handle >= 10,000 writes/s (WPS) per namespace, but indexing cannot currently keep up. This causes high query latency while performing bulk imports. When write throughput decreases (<= 100 per second) the indexer catches up, and queries will be fast.

Most use-cases do an initial bulk import, followed by queries with lower write throughput (<= 100 per second). For this use-case, it's not a problem. We are actively working to improve this limitation.

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.

MetricMax seen in productionProduction limits (current)Production limits (soon)
Max vectors (global)5.5B+
Max vectors (per namespace)100M+
Number of namespaces2M+Unlimited
Max dimensions10,752
Max inactive time in cacheContact us if you need custom
Write rate (global)10,000
Write rate (per namespace)10,000 10,000 vec/s100,000 vec/s
Max write batch rate (per namespace)1 batch/s
QPS (global)>100 QPS
Max QPS (per namespace)~20 QPS100+ QPS10,000 QPS
Max attribute value64 KiB1 MiB
Max attribute name length128128128
Max attributes per vector256256256
Max namespace name length128128128
Email us
© 2024 turbopuffer Inc.
Privacy PolicyTerms of service
SOC2 Type 1 certified