Vearch
A cloud-native, distributed vector database for similarity search over embedding vectors in AI applications, with hybrid vector and scalar filtering search, replication and elastic scale-out, and client SDKs for Python, Go, Java and Rust.
Vearch is a cloud-native, distributed vector database for efficient similarity search over embedding vectors. It combines vector search with scalar filtering in one hybrid query, retrieves from millions of objects in milliseconds, and scales out elastically with replication for reliability.
A Master component handles schema management and cluster-level metadata, a Router exposes a RESTful API for upsert, delete, search and query along with request routing and result merging, and PartitionServer nodes host document partitions with Raft-based replication. It integrates as a vector store or memory backend for LangChain, LlamaIndex and their Go and Java equivalents, and has been used to build large-scale visual search systems indexing billions of images.
It deploys through Helm on Kubernetes, Docker Compose in standalone or cluster mode, plain Docker, or compiled from source, and offers client SDKs for Python, Go, Java and Rust.