# tantaraio/voy

🕸️🦀 A WASM vector similarity search written in Rust

Repository: https://github.com/tantaraio/voy
Canonical: https://ross.abutalabs.com/products/voy
Homepage: https://www.npmjs.com/package/voy-search
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: k-d-tree, nearest-neighbor-search, similarity-search, vector-search, wasm, webassembly, rust, wasm-pack
Last push: 2023-09-20T18:14:53+00:00

## Health v2 (maintenance only)
Score: 31/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 93
- inputs: {"age_days": 1314, "days_push": 1078, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1070, forks 41 (observed 2026-08-28T04:03:27.903423+00:00)

## What it is
Voy is a tiny vector similarity search engine written in Rust and compiled to WebAssembly, using a k-d tree index for fast nearest-neighbor search. It ships as an npm package (voy-search) designed for browser and CDN edge deployments, with a portable, resumable embeddings index.

## Use cases
- run vector similarity search entirely in the browser
- add semantic search to a static site without a backend
- deploy a small vector search index to CDN edge servers
- search embeddings client-side with Web Workers
- build a portable embeddings index that can be shared and reloaded
- do nearest-neighbor search over embeddings in JavaScript via WASM

## When to choose
- you need lightweight in-browser or edge vector search without a server
- bundle size matters and you want a tree-shakable ~75KB library
- your dataset is small-to-medium and a k-d tree index suffices
- you want a portable index file you can rebuild or move anywhere

## When to avoid
- you need a stable API - Voy is pre-1.0 with possible breaking changes
- you need large-scale vector search with frequent index updates without full rebuilds
- you need built-in text embedding generation - Voy relies on external libraries like transformers.js
- you need server-side ANN at scale (consider dedicated vector databases like FAISS, Qdrant, or pgvector)

## Facets
- artifact type: library
- maturity: experimental
- function: search-engine, vector-database, wasm
- domain: machine-learning, web-development, developer-tools
- platform: browser, wasm, rust
- tags: k-d-tree, nearest-neighbor-search, similarity-search, vector-search, embeddings, client-side-search, edge-deployment, rust-wasm, search, nodejs, web-server

## Member repositories
- tantaraio/voy (main) score 31

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:27.903423+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:54:09.339764+00:00, confidence not recorded.
  - readme: https://github.com/tantaraio/voy (fetched 2026-08-28T04:03:27.903423+00:00, sha e5798df9c8ea)
- Data as of 2026-08-30T08:39:29.467469+00:00.