All Tools
N
Vector DBFreeOpen Source
NGT
Graph-and-tree ANN search for high-dimensional data
Apache-2.0
ABOUT
Exact nearest neighbor search collapses at high dimension and large corpus size. NGT builds neighborhood graphs and trees so you can run approximate k-NN over embedding vectors with better recall-latency tradeoffs than brute force, without standing up a full vector database.
INSTALL
pip install ngtINTEGRATION GUIDE
1. Index high-dimensional embeddings for fast approximate nearest neighbor lookup
2. Power similarity search in a custom retrieval stack without a hosted vector DB
3. Benchmark graph-based ANN against HNSW for large-scale embedding search
4. Add Python NGT indexes to a local RAG prototype before moving to a managed store
TAGS
annvector-searchembeddingscpluspluspythonnearest-neighboropen-source