habedi/multi-vector-hnsw

A Java implementation of HNSW with multi-vector search support

22
/ 100
Experimental

This is a Java library for developers who build applications requiring lightning-fast similarity searches on complex data. It indexes objects represented by multiple numerical descriptions (vectors) and quickly finds the most similar items. Developers can use this to power features like document search, multi-modal content recommendations, or hybrid data retrieval.

No commits in the last 6 months.

Use this if you are a Java developer building a system that needs to perform efficient approximate nearest neighbor searches where each item has several distinct numerical representations.

Not ideal if you are not a Java developer or if your search needs are simple and each item is represented by only a single vector.

information-retrieval similarity-search recommendation-engines multi-modal-ai data-indexing
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

10

Forks

Language

Java

License

Apache-2.0

Last pushed

Oct 14, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/habedi/multi-vector-hnsw"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.