We chose Milvus because it is very user-friendly. Without Milvus, our semantic search wouldn’t have been 10x smarter!

by Rahul Yadav, Principal Engineer at Tokopedia

Milvus drastically accelerated retrieval of millions of semantic vectors. We saw a near tenfold advancement compared to how things were when we were using other vector similarity search engines.

by Tingting Wang, NLP algorithm engineer at Sohu

We are looking forward to any future updates from Milvus that could help us further optimize the shopping experience for our users.

by Jessie Ji, Java Software Engineer at VOVA

The fast similarity search provided by Milvus allows ArtLens AI to match user uploaded photos with visually similar art from the CMA’s massive collection in seconds.

by Anna Faxon and Haley Kedziora, Digital Project Managers at Cleveland Museum of Art

Without a vector search engine like Milvus, similarity searches would not be feasible across the entire vector space.

by Elizabeth Edmiston, Senior Software Engineer at Lucidworks

The Milvus community is very active and has an abundance of resources available for its users. This helped us quickly developed a minimum viable product (MVP) and significantly lowered the development costs.

by Zhaoxing Li, Senior Engineer at Opera

Without Milvus, building our news aggregator wouldn't have been possible. The platform maintains consistent performance levels even in situations with unusually high concurrency.

by Jie Yuan, Recommendation Algorithm Architect at Xiaomi

Milvus’ flexibility makes it easy to optimize machine learning projects to better fit different scenarios.

by  Lang Wang, Senior Software Engineer at Kingsoft


  • supporting


    Milvus improves upon high-performance storage and index libraries such as Faiss and hnswlib, guaranteeing time and resource efficient query speeds. Using a variety of acceleration methods (e.g., CPU, SIMD, GPU, and FPGA), Milvus can retrieve vector data on trillion-scale datasets in milliseconds.

  • search

    On-demand Performance

    Milvus is cloud native and can be horizontally scaled with ease. The platform is capable of handling storage and computation at any scale, and has a microservice design that supports on-demand and automatic scaling.

  • deployment

    Built for Unstructured Data

    Milvus helps users focus on the semantic meaning of unstructured data rather than complexities such as sharding, data persistence, and load balancing. Milvus supports high-performance, hybrid search of vector and scalar data, opening up new possibilities for unstructured data processing.

  • storage


    Milvus is a graduate of the LF AI & Data Foundation's incubator program and has been adopted by 1,000+ organizations worldwide. The platform's vibrant open-source community welcomes contributions from everyone.

  • autoscaling


    Milvus supports multiple data types and provides comprehensive multi-language SDKs. Through the Python-ORM API, Milvus offers an integrated user experience across laptops, local clusters, and the cloud. A wealth of deployment and visualization tools are available to help users get Milvus up and running faster.

  • icon


  • 爱奇艺
  • 快手
  • 小米
  • line
  • moj
  • museum
  • Smart News
  • Compass
  • axa
  • Daily Hunt
  • Message Bird

Join the community

Milvus is open-sourced on GitHub.

Contributions and feedbacks are welcome!