vearch/vearch

Vearch is an open-source, cloud-native distributed vector database designed for AI-native applications. It enables efficient similarity search and scalar filtering on embedding vectors, crucial for tasks like visual search, recommendation systems, and RAG. With features like high performance, scalability through replication and elastic scaling, and a robust architecture comprising Master, Router, and PartitionServer components, Vearch provides a reliable and performant backend for modern AI workloads. It integrates seamlessly with popular AI frameworks like Langchain and LlamaIndex, offering comprehensive SDKs for various programming languages.