Crate crvecdb 

crvecdb

A fast vector database library with HNSW indexing for ARM64 and x86-64.

Features

• HNSW (Hierarchical Navigable Small World) indexing for fast approximate nearest neighbor search
• Multiple distance metrics: Cosine, Euclidean (L2), Dot Product
• Memory-mapped file storage for persistence
• Cross-platform SIMD acceleration (ARM NEON, x86 SSE/AVX2)
• Scales to millions of vectors

Example

use crvecdb::{Index, DistanceMetric};

// Create an in-memory index
let mut index = Index::builder(128)
    .metric(DistanceMetric::Cosine)
    .m(16)
    .ef_construction(200)
    .capacity(10_000)
    .build()
    .unwrap();

// Insert vectors
index.insert(1, &vec![0.1; 128]).unwrap();
index.insert(2, &vec![0.2; 128]).unwrap();

// Search for nearest neighbors
let results = index.search(&vec![0.15; 128], 10).unwrap();
for result in results {
    println!("ID: {}, Distance: {:.4}", result.id, result.distance);
}

Memory-Mapped Storage

use crvecdb::{Index, DistanceMetric};

// Create a persistent index
let mut index = Index::builder(768)
    .metric(DistanceMetric::DotProduct)
    .capacity(1_000_000)
    .build_mmap("/path/to/index.db")
    .unwrap();

// Data is automatically persisted via mmap
index.insert(1, &vec![0.1; 768]).unwrap();
index.flush().unwrap(); // Ensure durability

Structs

HnswConfig

HNSW algorithm configuration

Index

The main vector index type

IndexBuilder

Builder for creating indexes

IndexConfig

Complete index configuration

SearchResult

Search result with ID and distance

Enums

CrvecError

Error types for crvecdb operations

DistanceMetric

Distance metric selection

Type Aliases

InternalId

Internal index within storage (max ~4B vectors)

Result

Result type alias for crvecdb operations

VectorId

Unique identifier for a vector (user-provided)