Search iThub

Search tools, categories and collections. Type at least two characters.

Explore all tools
↑ ↓ to explore resultsEnter to open · Esc to close

Milvus

Open-source vector database for large-scale AI search

Visit website Save

Milvus is an Apache 2.0 vector database for searching text, image and multi-modal data, from the Milvus Lite quickstart to distributed clusters or Zilliz Cloud.

Pricing
Freemium
Deployment
Self-hosted
46,608stars4,326forks
22views
Updated

What is Milvus?

Milvus is a high-performance vector database built for scale. It powers AI applications by organizing and searching large amounts of unstructured data such as text, images and multi-modal information. It is written in Go and C++ and uses hardware acceleration for CPU and GPU.

According to the README, its fully distributed, Kubernetes-native architecture lets Milvus scale horizontally, handle tens of thousands of search queries on billions of vectors, and keep data fresh with real-time streaming updates. Milvus also supports a Standalone mode for single-machine deployment, and Milvus Lite is a lightweight version for quickstarts in Python installed with pip.

The project is under the LF AI & Data Foundation and is distributed under the Apache 2.0 license, with Zilliz as its major contributor. Milvus is also available as a fully managed service on Zilliz Cloud. The community uses Discord for questions and GitHub for issues and discussions.

Key capabilities

  • Vector search over text, image and multi-modal data.
  • Milvus Lite for a quickstart with pip in Python.
  • Standalone mode for single-machine deployment.
  • A distributed, Kubernetes-native architecture that scales horizontally.
  • A fully managed option on Zilliz Cloud.
Advertisements

Where Milvus fits

  • Vector search with filtering Vectors are stored together with scalar types such as integers, strings and JSON objects, so searches can use metadata filtering or hybrid search.
  • Index types The engine supports the major vector index types, including HNSW, IVF, FLAT, SCANN and DiskANN.
  • Hardware acceleration Hardware acceleration improves search performance, and GPU indexing is supported.
  • Independent scaling Compute and storage are separate, so query nodes scale for read-heavy loads and data nodes for write-heavy loads.
  • Multi-tenancy Isolation at database, collection, partition or partition key level lets one cluster handle hundreds to millions of tenants.
  • Hot and cold storage Frequently accessed data can sit in memory or on SSDs, while less-accessed data goes to cheaper storage.
  • Deployment modes Milvus Lite installs with pip for quickstarts, Standalone runs on one machine, and Zilliz Cloud offers a managed service.

Install

pip install -U pymilvus

Reviews of Milvus

No reviews yet

Be the first to share your experience with Milvus.

Loading review form…

Alternatives to Milvus

Compare Milvus with these tools →