Start embedded, move to a managed instance when you need it. Keep the same retrieval engine and client API; we handle provisioning, updates and monitoring.
Start with a 42-day sandbox, then manage collections, API keys, usage, and billing as your project grows.
PaveDB is open source: read the code, audit it, run your own — everything is at pavedb.org. And when you'd rather build than operate, we run it for you.
Small enough to embed. Complete enough to operate. PaveDB is an open-source (AGPL-3.0) retrieval database with document ingestion, source provenance, tenant controls and metrics. Compare logged text queries with reruns to see how results change as your collection evolves.
Provisioning, updates, and monitoring handled by Flowlexi. Backup coverage follows your plan.
Capacity for your project
Choose shared hosting or dedicated capacity according to your plan and workload.
Your keys, your data
Keep control of API credentials and export your collections. Import them into a compatible PaveDB instance, including one you operate yourself. Same database, a different operator.
