Turbopuffer

Turbopuffer is a proprietary, managed search engine from turbopuffer that offers vector, full-text and filtered search. It uses an object-storage-native design that caches data on SSD and in memory, and its documentation states higher write latency and higher cold-query latency as a trade-off of that approach. It is available only as a managed service, in multi-tenant or single-tenant form, or as a Bring Your Own Cloud deployment. No release versions or release dates are stated. Pricing is based on monthly minimums. The Launch plan has a $16 monthly minimum and the Scale plan a $256 monthly minimum. The Enterprise plan has a $4,096 monthly minimum plus usage. Because the service is managed and closed, there is no self-managed option, and buyers should weigh the latency trade-off the vendor describes against their workload before committing. Turbopuffer suits teams that want search over large collections of data without running the infrastructure themselves, and that can accept the stated trade-off in exchange for object-storage economics. It is also categorised as a search engine, so it may be compared with other search products as well as vector databases.