An in-process vector database that runs locally without a separate server, offering similarity search for applications that need minimal operational overhead. It supports vector embeddings, keyword full-text search, and hybrid retrieval within a library or desktop GUI interface.
Project overview
It embeds directly into application code and runs wherever code runs without server configuration, which addresses the operational overhead of deploying and managing an external database server.
Project type
RAG · AI Search · Vector Database
Use cases
Knowledge Q&A · Search & Research
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers, data teams, and AI engineers who need to add similarity or hybrid search directly to their application logic without external server dependencies.
Key capabilities
Enables native keyword-based full-text search to query string fields with natural-language or structured expressions.
Fuses vector similarity, full-text search, and structured filters in a single query for precise results.
Limitations and risks
The architecture is designed to run purely locally and in-process, so it is not designed to act as a standalone database server.
Minimum hardware requirements, GPU requirements, and telemetry practices are not documented.
Getting started
Install the library via standard package managers like pip. Once installed, import and use the system in code or through Zvec Studio.