What is PistaDB?

PistaDB is a lightweight embedded vector database written in pure C99 with zero external dependencies. It ships as a single shared library and stores everything in a single .pst file.

It is designed for local RAG pipelines, offline AI agents, privacy-sensitive applications, edge inference, and anywhere shipping a full vector database cluster is impractical — which, honestly, is most places.

PistaDB vs. cloud / server vector databases

	PistaDB	Cloud / Server Vector DB
Deployment	Copy a .dll / .so	Docker, Kubernetes, cloud subscriptions
Storage	One .pst file	Separate data + WAL + config + sidecar files
Privacy	All data stays local	Embeddings sent over the network
Memory	Configurable, minimal	GBs of JVM / runtime overhead
Dependencies	None (pure C99)	Dozens of packages
Latency	Sub-millisecond on a laptop	Network round-trips
Cost	Free forever (MIT)	Per-query or per-vector pricing

When PistaDB fits

• Desktop apps that need semantic search without phoning home.
• Mobile RAG (Android via JNI, iOS via Swift Package Manager).
• Edge / MCU inference (experimental ESP32 component).
• Embedded analytics inside an existing C/C++/Go/Rust application.
• Notebook prototyping before deciding whether a cluster is even needed.

When it doesn't

• Multi-tenant SaaS with thousands of concurrent writers — use a server DB.
• Datasets that don't fit on a single machine — PistaDB scales to ~10 M vectors per file, not 10 B.
• Workloads that need a managed control plane, replication, or distributed consensus.

Next steps

• Getting Started — install and run your first query in 60 seconds.
• Index Algorithms — pick the right index for your workload.
• Multi-Modal Retrieval — multiple vector fields per record + RRF hybrid search.
• Language Bindings — Python, Go, Rust, Swift, Kotlin, C#, and more.