OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Hora

Hora is an open-source Rust library of approximate nearest-neighbor search algorithms for finding similar items in high-dimensional numeric data. Developers can build indexes and search them through documented Rust, Python and JavaScript/WebAssembly interfaces.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Hora — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Hora when your application already has numeric vectors and needs an index for finding similar items, such as image features or recommendation data. It provides retrieval algorithms rather than embedding generation or a complete hosted search application.

Who it is for

  • Rust developers building vector-similarity retrieval into an application or service.
  • Machine-learning and search developers with numeric vectors or embeddings who need nearest-neighbor lookup.
  • Teams exploring image search or recommendation retrieval using approximate-neighbor indexes.

Documented uses

  • Index numeric vectors for similarity retrieval.
  • Search for the nearest items to a query vector.
  • Choose among documented index types and distance measures for a retrieval component.
  • Integrate approximate-neighbor search into a Rust, Python or JavaScript/WebAssembly project.
Before you choose

Requirements & limits

The project is available as a public GitHub repository under the Apache-2.0 license. Its README documents package installation for Rust, Python and JavaScript/WebAssembly.

  • The Rust README provides concrete steps for building an HNSW index and querying it; the demonstrated example uses Euclidean distance.
  • The README lists several index choices, but marks the random-projection-tree option as work in progress.
  • Python and JavaScript/WebAssembly have installation instructions; Java is listed as supported, while Go, Ruby, Swift, R and Julia are marked work in progress.
  • The project describes SIMD acceleration and multithreading, but workload-specific performance should not be inferred from those claims.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Build an HNSW index and search vectors in Rust

Follow the documented Rust example to add fixed-dimension vectors to an HNSW index and retrieve nearest neighbors.

Before you start

  • Rust project with Hora added as a dependency in Cargo.toml.
  • Numeric vectors with a consistent dimension; the example uses f32 values and a 64-dimensional index.

Steps

  1. Add hora = "0.1.1" under [dependencies] in Cargo.toml.

  2. Create an HNSW index for the vector dimension using default HNSW parameters.

  3. Add each sample vector with its associated item identifier.

  4. Build the index using Euclidean distance, then call search with a query vector and desired neighbor count.

Expected outcome

An HNSW index that returns approximate nearest neighbors for a query vector, as illustrated by the Rust example.

Guide limitations

  • The example uses Euclidean distance and default HNSW parameters; it does not explain parameter tuning or persistence.
  • The README example is illustrative and is not a performance or compatibility guarantee.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
30 September 2026
Original website title
hora-search/hora · Official homepage title · retained in its source language
How we review sourcesSuggest a correction