# arroy

> Rust library for low-latency on-device vector similarity search using random projection trees and LMDB storage, enabling efficient ANN on edge devices. Supports concurrent multi-process access for real-time AI apps. Ideal for IoT and embedded systems vs cloud alternatives like Qdrant.

- **URL:** https://github.com/meilisearch/arroy
- **Category:** Embedded And Edge Vector Databases
- **Tags:** Open Source, Vector Embeddings, Similarity Search, Rust Lang, Edge Ai
- **Updated:** 2026-04-30 11:38
- **Canonical page:** https://vectordb.ever.works/fr/items/arroy

## Details

## Overview
Arroy provides efficient on-device vector search with low-latency ANN using LMDB backend.

## Features
- Low-latency ANN search with random projection trees
- LMDB for concurrent read/write access
- Configurable n_trees and search_k for latency/accuracy trade-off
- Multi-threaded queries

## Use Cases
- IoT real-time similarity search
- On-device AI embeddings management
- CV inference on edge hardware

## Comparisons
- On-device low-latency vs cloud Qdrant servers
- Lower memory than Annoy, persistent unlike in-memory only libs

## Pricing
Free open-source (MIT).

---

_This Markdown mirror is generated by Ever Works for AI agents. The canonical HTML page is at https://vectordb.ever.works/fr/items/arroy._