unum-cloud/USearch

Fast Open-Source Search & Clustering engine × for Vectors & Arbitrary Objects × in C++, C, Python, JavaScript, Rust, Java, Objective-C, Swift, C#, GoLang, and Wolfram 🔍

What it solves

USearch is a high-performance similarity search and clustering engine designed to be smaller, faster, and more portable than traditional vector search libraries like FAISS. It addresses the need for an efficient way to find the nearest neighbors of vectors (embeddings) across a wide variety of programming languages and hardware platforms without requiring heavy dependencies.

How it works

USearch implements the HNSW (Hierarchical Navigable Small World) algorithm for approximate nearest neighbor search. It optimizes performance using SIMD (Single Instruction, Multiple Data) and JIT compilation for user-defined metrics. To reduce memory footprints, it supports a wide range of numeric types (downcasting and quantization), including half-precision (bf16, f16) and quarter-precision formats. It also allows indexes to be served directly from disk via memory-mapping, avoiding the need to load the entire index into RAM.

Who it’s for

It is intended for developers building AI applications that require semantic search, genomics or chemistry applications (using binary Tanimoto/Sorensen coefficients), and those needing to integrate vector search into existing databases or cross-platform applications (iOS, Android, WebAssembly).

Highlights

  • Extreme Performance: Claims up to 10x faster indexing than FAISS and significantly faster brute-force search.
  • Broad Compatibility: Native bindings for 10+ languages including Python, Rust, C++, Go, and Java.
  • Memory Efficient: Supports uint40_t for indexing up to 1 trillion entries and multiple low-precision data types to save RAM.
  • Flexible Metrics: Supports arbitrary user-defined metrics via Numba, Cppyy, or PeachPy.
  • Disk-Based Serving: Ability to view large indexes from disk without loading them into RAM, potentially reducing cloud infrastructure costs.
  • Integrated Clustering: Built-in K-Nearest Neighbors clustering and sub-clustering capabilities.

Related

  • Project
    facebookresearch/faiss

    A library for efficient similarity search and clustering of dense vectors, capable of handling billion-scale datasets using CPU and GPU acceleration.

  • Project
    NVIDIA/cuvs

    A GPU-accelerated library for vector search and clustering, providing high-performance approximate nearest neighbors implementations for AI and data mining.

  • Project
    MinishLab/vicinity

    Vicinity is a lightweight Python library that provides a unified, low‑dependency interface for nearest‑neighbour search over vector embeddings. It supports many back‑ends (FAISS, HNSW, Annoy, etc.), offers easy serialization (local disk or Hugging Face Hub), and includes a simple evaluation tool for measuring speed and recall.

  • Project
    RediSearch/RediSearch

    A Redis module that adds full-text search, secondary indexing, and vector similarity search to Redis, enabling high-performance querying of database records.

  • Project
    SeekStorm/SeekStorm

    A high-performance Rust-based search engine providing sub-millisecond native vector and lexical search, available as an in-process library or a multi-tenancy server.