Crate ruvector_core 

Ruvector Core

High-performance Rust-native vector database with HNSW indexing and SIMD-optimized operations.

Working Features (Tested & Benchmarked)

• HNSW Indexing: Approximate nearest neighbor search with O(log n) complexity
• SIMD Distance: SimSIMD-powered distance calculations (~16M ops/sec for 512-dim)
• Quantization: Scalar (4x) and binary (32x) compression with distance support
• Persistence: REDB-based storage with config persistence
• Search: ~2.5K queries/sec on 10K vectors (benchmarked)

⚠️ Experimental/Incomplete Features - READ BEFORE USE

• AgenticDB: ⚠️⚠️⚠️ CRITICAL WARNING ⚠️⚠️⚠️
  • Uses PLACEHOLDER hash-based embeddings, NOT real semantic embeddings
  • “dog” and “cat” will NOT be similar (different characters)
  • “dog” and “god” WILL be similar (same characters) - This is wrong!
  • MUST integrate real embedding model for production (ONNX, Candle, or API)
  • See agenticdb module docs and /examples/onnx-embeddings for integration
• Advanced Features: Conformal prediction, hybrid search - functional but less tested

What This Is NOT

• This is NOT a complete RAG solution - you need external embedding models
• Examples use mock embeddings for demonstration only

Re-exports

pub use advanced_features::ConformalConfig;

pub use advanced_features::ConformalPredictor;

pub use advanced_features::EnhancedPQ;

pub use advanced_features::FilterExpression;

pub use advanced_features::FilterStrategy;

pub use advanced_features::FilteredSearch;

pub use advanced_features::HybridConfig;

pub use advanced_features::HybridSearch;

pub use advanced_features::MMRConfig;

pub use advanced_features::MMRSearch;

pub use advanced_features::PQConfig;

pub use advanced_features::PredictionSet;

pub use advanced_features::BM25;

pub use agenticdb::AgenticDB;

pub use embeddings::EmbeddingProvider;

pub use embeddings::HashEmbedding;

pub use embeddings::BoxedEmbeddingProvider;

pub use embeddings::ApiEmbedding;

pub use error::Result;

pub use error::RuvectorError;

pub use types::DistanceMetric;

pub use types::SearchQuery;

pub use types::SearchResult;

pub use types::VectorEntry;

pub use types::VectorId;

pub use vector_db::VectorDB;

Modules

advanced

Advanced techniques: hypergraphs, learned indexes, neural hashing, TDA (Phase 6)

advanced_features

Advanced Features for Ruvector

agenticdb

AgenticDB API Compatibility Layer

arena

Arena allocator for batch operations

cache_optimized

Cache-optimized data structures using Structure-of-Arrays (SoA) layout

distance

SIMD-optimized distance metrics Uses SimSIMD when available (native), falls back to pure Rust for WASM

embeddings

Text Embedding Providers

error

Error types for Ruvector

index

Index structures for efficient vector search

lockfree

Lock-free data structures for high-concurrency operations

quantization

Quantization techniques for memory compression

simd_intrinsics

Custom SIMD intrinsics for performance-critical operations

storage

Storage layer with redb for metadata and memory-mapped vectors

types

Core types and data structures

vector_db

Main VectorDB interface