All Tools
Q
Vector DBFreemiumOpen Source
QDRANT
Vector database built for the next generation of AI applications
Apache-2.0
ABOUT
Most vector databases sacrifice either performance or flexibility. Qdrant is written in Rust for maximum throughput and offers payload filtering (filter by metadata before or after vector search), named vectors (multiple embeddings per record), and on-disk indexing for large datasets that don't fit in RAM. Self-host for free or use Qdrant Cloud for managed hosting.
INSTALL
npm install @qdrant/js-client-restINTEGRATION GUIDE
1. Build a semantic document search engine that filters by metadata (date, author, category) alongside vector similarity
2. Store multiple embedding models per document (e.g. dense + sparse) for hybrid retrieval
3. Run a RAG pipeline where the knowledge base is too large for in-memory vector stores
4. Power recommendation systems that combine user preference vectors with catalog item vectors
5. Build a code search tool over a large codebase using code-specific embedding models
TAGS
rustpythontypescriptvector-searchembeddingsraghybrid-search