HNSW (Rust)

Category: SDKs & Libraries
Brand: rust-cv
Website/Source: https://github.com/rust-cv/hnsw
License: MIT
Slug: hnsw-rust

Description

A Rust implementation of the Hierarchical Navigable Small World (HNSW) approximate nearest neighbor (ANN) search algorithm, designed for fast, memory-safe vector search in Rust-based AI, retrieval, and search systems.

Features

• Hierarchical Navigable Small World (HNSW) graph implementation for efficient approximate nearest neighbor search.
• Fast ANN search optimized for high-performance vector similarity queries.
• Rust-native library leveraging Rust’s safety and performance guarantees.
• serde integration (optional feature flag): enable the serde feature to serialize and deserialize HNSW data structures.
• Examples and benches included in the repository (examples, benches folders) to demonstrate usage patterns and performance.
• Tests and implementation notes (tests, implementation.md) documenting and validating the algorithm behavior.
• No-std compatibility checks via ensure_no_std directory, indicating attention to usage in constrained environments (where applicable).

Pricing

• Open-source under the MIT License (no usage fee).