vearch/vearch

Vearch is an open-source, cloud-native distributed vector database optimized for AI-native applications. It enables efficient similarity search of embedding vectors, supporting both vector search and scalar filtering for complex queries. With its high-performance retrieval capabilities, Vearch can search millions of objects in milliseconds, offering robust scalability and reliability through replication and elastic scaling. It integrates seamlessly with popular AI frameworks like Langchain and LlamaIndex, serving as a powerful memory backend for various AI-driven use cases, including visual search systems.