Zilliz

The product is a vector lakebase and managed vector database for storing and searching vectors, text, JSON and geospatial data at hundred-billion scale, supporting real-time serving, iterative discovery and batch analytics on a single source of truth. It is for developers, data teams and IT teams in businesses that build AI applications requiring similarity search and retrieval. It is delivered as a fully managed cloud service with BYOC options and as an open-source vector database, positioned as an evolution beyond traditional vector databases to unify serving and analytics.

Key features

• Real-time vector search
• Iterative discovery at lake scale
• Batch analytics on single source
• Billion-scale vector similarity search
• Hybrid retrieval with filtering
• Reranking for multi-modal queries
• BM25 full-text search
• Vector and text search
• JSON and geospatial search
• Hot cache with on-demand compute
• Lake-native storage on S3
• Handles 100B+ entities and 10K+ QPS