VectorLite

A lightweight vector database with built-in embeddings generation, search and retrieval capabilities for AI agents

Features

• Two Index Types: Flat and HNSW indexing
• ChromaDB-like API: Simple client interface with collections
• Multiple Similarity Metrics: Cosine, Euclidean, Manhattan, Dot Product
• Built-in Embeddings: Uses all-MiniLM-L6-v2 model
• Memory-only: No persistence, all data in RAM

Quick Start

use vectorlite::{VectorLiteClient, EmbeddingGenerator, IndexType, SimilarityMetric};

// Create client with embedding function
let mut client = VectorLiteClient::new(Box::new(EmbeddingGenerator::new()?));

// Create collection
client.create_collection("documents", IndexType::Flat)?;

// Add text (auto-generates embedding and ID)
let id = client.add_text_to_collection("documents", "Hello world")?;

// Search
let collection = client.get_collection("documents").unwrap();
let results = collection.search_text("hello", 5, SimilarityMetric::Cosine)?;

Index Types

• FlatIndex: O(n) search, good for small datasets
• HNSWIndex: O(log n) search, good for large datasets

Similarity Metrics

• SimilarityMetric::Cosine - Cosine similarity (default)
• SimilarityMetric::Euclidean - Euclidean distance
• SimilarityMetric::Manhattan - Manhattan distance
• SimilarityMetric::DotProduct - Dot product

HNSW Configuration

Build with different performance profiles:

cargo build                           # Default (balanced)
cargo build --features memory-optimized  # Lower memory
cargo build --features high-accuracy     # Higher accuracy

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.