Zvec
Open-source, in-process vector database by Alibaba, positioned as the SQLite of vector databases. Delivers production-grade, low-latency similarity search with minimal setup, achieving >8,000 QPS on VectorDBBench. This is an open-source (OSS) solution released under Apache 2.0 license.
About this tool
Overview
Zvec is a lightweight, lightning-fast, in-process vector database released by Alibaba Tongyi Lab in February 2026. It runs as a library inside your application with no external service or daemon required, making it the SQLite of vector databases.
Key Features
- In-Process Architecture: Runs wherever your code runs — notebooks, servers, CLI tools, or edge devices
- Dense + Sparse Vectors: Work with both dense and sparse embeddings, with native support for multi-vector queries
- Hybrid Search: Combine semantic similarity with structured filters for precise results
- Built on Proxima: Powered by Alibaba's production-grade, battle-tested vector search engine
- WAL Support: Provides Write-Ahead Logging for data durability
- RAG Ready: Full CRUD operations, schema evolution, multi-vector retrieval, built-in reranking (weighted fusion and RRF)
Performance
Zvec delivers >8,000 QPS on VectorDBBench with the Cohere 10M dataset, achieving more than 2× the performance of previous leaderboard leaders while reducing index build time.
Platform Support
- Python 3.10 to 3.12
- Linux x86_64, Linux ARM64, macOS ARM64
- Install with:
pip install zvec
Use Cases
- Retrieval Augmented Generation (RAG)
- Semantic search on edge devices
- Agent workloads requiring local execution
- Mobile and constrained hardware deployments
Pricing
Free and open-source under Apache 2.0 license. No licensing costs.
Loading more......
Information
Categories
Tags
Similar Products
6 result(s)Orama is a lightweight search engine that supports vector and hybrid search functionalities, suitable for browser, server, or edge environments.
Lightweight Python package for storing and retrieving text using chunking, embeddings, and vector search. Powers AI features in Kagi Search with low latency and small memory footprint. This is an OSS library.
Open-source AI search platform combining vector search, keyword retrieval, structured filtering, and ML ranking. Powers applications at Spotify, Yahoo, and Wix with sub-100ms response times. This is an OSS platform under Apache 2.0 with managed cloud option.
Qdrant is an open‑source vector database designed for high‑performance similarity search and AI applications such as RAG, recommendation systems, advanced semantic search, anomaly detection, and AI agents. It provides scalable storage and retrieval of vector embeddings with features like filtering, hybrid search, and production‑grade APIs for integrating with machine learning workloads.
Qdrant Edge is a private beta offering of Qdrant optimized for edge and on-device deployments, enabling low-latency vector search and AI capabilities closer to where data is generated.
tinyvector is a minimal vector database / ANN engine focused on simplicity and compact implementation for educational and small-scale similarity search uses.