Milvus is a cloud-native vector database used to manage embedded vectors. These vectors are generated by machine learning and neural networks models. The product advances the basic capabilities of approximate nearest neighbor (ANN) search libraries and facilitates on-demand scalability.

Also, it helps simplify unstructured data management while delivering a consistent experience in various deployment environments. It has been adopted by more than 1,000 companies globally in different use cases. Milvus 2.0 is the latest version which is cloud-native first, supporting unified batch and stream processing.

Project Background

  • Database: Milvus
  • Author: Jianguo Wang, Xiaomeng Yi, Rentong Guo, Hai Jin, Peng Xu, Shengjun Li
  • Initial Release: October 2019
  • Type: Vector Database
  • License: Apache License 2.0
  • Contains: Milvus, etcd metadata engine, MinIO storage engine
  • Language: Python, Go, C++
  • GitHub: /milvus-io/milvus/ has 8.4k stars, 62 releases, and 164 contributors 
  • Runs On: macOS, Linux, Windows with WSL 2 enabled
  • GitHub Discussions: /milvus-io/milvus/discussions
  • Stackflow: /milvus
  • Twitter: /milvusio  


  • Image similarity search
  • Video similarity search
  • Audio similarity search
  • Molecular similarity search
  • Recommender system
  • Question answering system
  • DNA sequence classification  
Scroll to Top