Turbovec

Features

Vector Embedding Indexes - Provides a high-performance index for storing and retrieving high-dimensional embeddings via similarity search.
Vector Memory Stores - Functions as a pluggable vector store designed to replace in-memory storage in LLM orchestration frameworks.
Orchestration Store Integrations - Provides a pluggable vector store implementation that integrates directly into LLM orchestration frameworks.
Random Rotation Compression - Uses random orthogonal matrices to distribute variance and improve the accuracy of quantized vector searches.
Vector Search Indexes - Provides a quantized vector index for high-performance similarity search using SIMD kernels.
Vector Databases - A memory-efficient vector database written in Rust with Python bindings for embedding storage and retrieval.
Memory-Optimized Storage - Reduces the memory footprint of vector indices by using low-bit representations of embeddings.
Vector Quantization - Implements vector quantization using random rotation and scalar quantization to reduce memory footprints.
Coordinate-Specific Scalar Quantization - Compresses high-dimensional embeddings using coordinate-specific shift and scale values for memory efficiency.
Data-Oblivious Quantization - Reduces memory footprint by converting floating-point vectors into integers using data-oblivious quantization.
High-Dimensional Vector Compressors - Implements a vector compressor that uses random rotation and data-oblivious quantization to reduce memory usage.
Vector Search - Provides a high-performance implementation of vector similarity search for high-dimensional embeddings.
SIMD-Accelerated Arithmetic - Accelerates nearest neighbor retrieval using SIMD-accelerated kernels for maximum throughput.
SIMD Accelerated Searchers - Utilizes SIMD-accelerated kernels to perform high-throughput distance calculations and memory scanning.
Calibration Parameters - Fits empirical data distributions to coordinate-specific shift and scale values during data ingestion.
Stable Index Mappings - Maintains stable mappings between vector IDs and physical locations to support deletions without rebuilding the corpus.
Real-Time Vector Ingestion - Enables adding new vectors to the index in real-time without requiring full corpus rebuilding.
Bitmask Filtering - Implements high-performance result filtering using bitmasks to isolate document subsets during retrieval.
Persistent Identifier Management - Assigns persistent IDs to vectors to maintain index consistency after entries are deleted.
Filtered Similarity Searches - Restricts vector similarity search results to specific document subsets using bitmask filtering.
Kernel-Level Filter Optimizations - Optimizes search performance by short-circuiting blocks of disallowed vectors within the processing kernel.

Open-source alternatives to Turbovec

Similar open-source projects, ranked by how many features they share with Turbovec.

lancedb/lancedb
lancedb/lancedb
9,031View on GitHub
LanceDB is a vector database and columnar data store designed to function as a versioned dataset manager and vector search engine. It serves as a high-performance backend for indexing and retrieving high-dimensional embeddings, providing the foundation for machine learning data pipelines. The system distinguishes itself through a combination of cloud-native object storage and immutable version tracking, allowing for data time-travel and reproducible AI experiments. It integrates hybrid search capabilities, merging dense vector similarity with BM25 full-text search and SQL-like scalar filters
HTMLapproximate-nearest-neighbor-searchimage-searchnearest-neighbor-search
View on GitHub9,031
tensorchord/pgvecto.rs
tensorchord/pgvecto.rs
2,175View on GitHub
pgvecto.rs is a database extension that integrates high-dimensional vector search capabilities directly into PostgreSQL. It functions as a specialized engine for storing and retrieving embeddings, allowing relational databases to perform similarity searches alongside traditional structured data queries. The extension distinguishes itself through hardware-aware execution strategies that maximize performance. It performs runtime analysis of the host machine to utilize specific processor instruction sets for accelerated mathematical operations. To manage memory efficiently, it employs quantizati
Rustchatgptfaissgpt
View on GitHub2,175
unum-cloud/usearch
unum-cloud/USearch
3,888View on GitHub
USearch is a high-performance vector similarity search engine and approximate nearest neighbor index designed for dense embeddings. It functions as a low-level vector database core and high-dimensional vector indexer, providing the primitives necessary to store and retrieve vectors across massive datasets. The engine distinguishes itself through hardware-level SIMD acceleration for distance kernels and a proximity-graph indexing system that enables fast retrieval across billions of vectors. It supports multi-precision vector quantization to balance memory usage and accuracy, and utilizes memo
C++approximate-nearest-neighbor-searchclusteringdatabase
View on GitHub3,888
redis/go-redis
redis/go-redis
22,159View on GitHub
This project is a feature-rich Go client library designed for interacting with Redis. It serves as a comprehensive interface for managing remote data stores, enabling developers to execute standard database commands, handle complex data structures, and perform asynchronous operations within Go applications. The library distinguishes itself through its support for advanced Redis capabilities, including connection pooling, pipelining, and transactional integrity. It provides specialized primitives for managing distributed clusters, including automated topology updates and request routing to sha
Gogogolangredis
View on GitHub22,159

See all 30 alternatives to Turbovec

RyanCodraiturbovec

Features

Open-source alternatives to Turbovec

lancedb/lancedb

tensorchord/pgvecto.rs

unum-cloud/USearch

redis/go-redis

Star history

Open-source alternatives to Turbovec

lancedb/lancedb

tensorchord/pgvecto.rs

unum-cloud/USearch

redis/go-redis